Supreme Court of the United States

No. 18-966

IN THE Supreme Court of the United States

DEPdARTMENT OF COMMERCE, ET AL., Petitioners, —v.—

STATE OF NEW YORK, ET AL., Respondents.

ON WRIT OF CERTIORARI BEFORE JUDGMENT TO THE UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT

JOINT APPENDIX (VOLUME 2)

Noel J. Francisco DALE E. HO* Solicitor General American Civil Liberties Department of Justice Union Foundation Washington, D.C. 20530-0001 125 Broad Street [email protected] New York, NY 10004-2400 (202) 514-2217 [email protected] (212) 549-2693 LETITIA JAMES Attorney General State of New York BARBARA D. UNDERWOOD* Solicitor General 28 Liberty Street New York, NY 10005-1400 [email protected] (212) 416-8016 Counsel of Record *Counsels of Record for Petitioners for Respondents

PETITION FOR A WRIT OF CERTIORARI FILED: JAN. 25, 2019 CERTIORARI GRANTED: FEB. 15, 2019

TABLE OF CONTENTS PAGE

Volume 2 of 4

Excerpts from Subjects Planned for the 2020 Census and American Community Survey (Mar. 2017) (A.R. 194–270) ...... 160

Emails between Kris Kobach and Wendy Teramoto (July 24, 2017) (A.R. 763–764) ...... 185

Letter from Vanita Gupta to Wilbur Ross (Jan. 4, 2018) (A.R. 773–775) ...... 188

Letter from Leadership Conference to Wilbur Ross (Jan. 10, 2018) (A.R. 798–803) ...... 193

Letter from Vincent Barabba, et al. to Wilbur Ross (Jan. 26, 2018) (A.R. 1057–1058) ...... 203

Letter from Jeff Landry to Wilbur Ross (Feb. 8, 2018) (A.R. 1079–1080) ...... 207

Letter from Eric Schneiderman, et al. to Wilbur Ross (Feb. 12, 2018) (A.R. 1090–1101) ...... 211

Summary of Conversation between Wilbur Ross and Hermann Habermann, Former Deputy Director and COO of Census Bureau (A.R. 1259–1260) ...... 235

Summary of Conversation between Secretary Ross and Christine Pierce, SVP of Data Science, Nielsen (A.R. 1276) ...... 238 i

PAGE

Emails between Wilbur Ross and Earl Comstock (Sept. 1, 2017) (A.R. 2424) ...... 241

Email from Earl Comstock to Wendy Teramoto (Sept. 16, 2017) (A.R. 2458) ...... 243

Email from Earl Comstock to Wilbur Ross (Mar. 10, 2017) (A.R. 2521–2523) ...... 245

Email from Wilbur Ross to Peter Davidson (Sept. 19, 2017) (A.R. 2528) ...... 251

Emails between Wendy Teramoto, Danielle Cutrona, and John Gore (Sept. 18, 2017) (A.R. 2636) ...... 252

Emails between Danielle Cutrona, Wendy Teramoto, and John Gore (Sept. 18, 2017) (A.R. 2651–2652) ...... 254

Email from John Abowd to John Philips, et al. (Feb. 20, 2018) (A.R. 2950–2953) ...... 257

Email from Ron Jarmin to Karen Kelley (Dec. 22, 2017) (A.R. 3289) ...... 264

Email from Ron Jarmin to Karen Kelley (Feb. 6, 2018) (A.R. 3460) ...... 266

Draft Letter from Wilbur Ross to Senator Harris (Jan. 2018) (A.R. 3560–3561) ...... 267

Letter from Civil Rights Legal Groups to Wilbur Ross (Mar. 22, 2018) (A.R. 3605–3606) ...... 269 ii

PAGE

Emails between David Langdon, Lisa Blumerman, Burton Reist, et al. (May 24, 2017) (A.R. 3702–3704) ...... 273

Email from Earl Comstock to Wilbur Ross (May 2, 2017) (A.R. 3710) ...... 276

Talking Points re: Citizenship Question Request and Process of Adding Content to Decennial Census (A.R. 3890–3891) ...... 277

Emails between Wilbur Ross and Earl Comstock (Aug. 10, 2017) (A.R. 3984) ...... 281

Email from Earl Comstock to Wilbur Ross (Aug. 8, 2017) (A.R. 4004) ...... 283

Excerpts from Submission of the 2020 Census and American Community Survey Questions to Congress (Feb. 2018) (A.R. 4802–4826) ...... 284

Memorandum from John Abowd to Ron Jarmin (Jan. 3, 2018) (A.R. 5473–5475) ...... 290

Email from Ron Jarmin to Arthur Gary (Jan. 3, 2018) (A.R. 5489-5491) ...... 296

Alternative Sources of Citizenship Data for the 2020 Census (Dec. 22, 2017) (A.R. 5500–5511) ...... 301

iii

PAGE

Summary of the Process for Using Administrative Records to Add Citizenship to the 2020 Census Microdata and Produce Block-level Tabulations from the Enhanced Data (Jan. 19, 2018) (A.R. 6415) ...... 319

Questions Assigned for the Census Memorandum (A.R. 7644–7645) ...... 322

Excerpts from Challenges and Experiences Adapting Differentially Private Mechanisms to the 2020 Census (2018) (A.R. 8912–8953) ...... 324

Email from Christa Jones to Sahra Park-Su (Feb. 24, 2018) (A.R. 9190) ...... 347

Emails between John Abowd and Mark Asiala (Jan. 29, 2018) (A.R. 9205–9207) ...... 348

Emails between Mark Asiala and John Abowd (Jan. 29, 2018) (A.R. 9218–9220) ...... 349

Email from John Abowd to Ron Jarmin (Jan. 29, 2018) (A.R. 9221) ...... 357

Emails between John Abowd, Mark Asiala, et al. (Jan. 29, 2018) (A.R. 9222–9224) ...... 359

Emails between Mark Asiala, John Abowd, et al. (Jan. 29, 2018) (A.R. 9225–9226) ...... 361

Memorandum from Center for Survey Measurement to Associate Directorate for Research and Methodology (Sept. 20, 2017) (A.R. 10386–10393) ...... 365 iv

PAGE

Checklist for Disclosure Avoidance Officers (Jan. 19, 2018) (A.R. 10733–10734) ...... 379

Questions on the Jan. 19 Draft Census Memo on the DOJ Citizenship Question Request (A.R. 10895–10902) ...... 381

Spreadsheet re: Question Assignments (A.R. 10950) ...... 395

Email from Matt Fidel, Religious Action Center of Reform Judaism to Wilbur Ross (Feb. 15, 2018) (A.R. 1122–1123) ...... 397

Emails between Earl Comstock and James Uthmeier (Aug. 11, 2017) (A.R. 11343–11345) ...... 401

Memorandum from John Abowd to Ron Jarmin (Dec. 22, 2017) (A.R. 11646–11648) ...... 404

Emails between Wilbur Ross and Earl Comstock (Aug. 10, 2017) (A.R. 12476) ...... 410

Emails between Earl Comstock, Wendy Teramoto, and Wilbur Ross (Aug. 10, 2017) (A.R. 12755–12756) ...... 412

Email from Christa Jones to Sahra Park-Su (Feb. 24, 2018) (A.R. 13023–13024) ...... 416

Emails between Karen Kelley, Israel Hernandez, Sahra Park-Su, et al. (Sept. 11, 2017) ...... 418

Joint Stipulations ...... 421 v

PAGE

Declaration of Steven Choi (Oct. 26, 2018) ...... 429

Supplemental Declaration of Steven Choi (Nov. 2, 2018) ...... 441

Excerpts from Declaration of Dr. Jennifer Van Hook ...... 448

Affidavit of Susan Brower (Oct. 25, 2018) ...... 462

Declaration of George Escobar (Oct. 26, 2018) ...... 466

Affidavit of Marchelle Franklin (Oct. 25, 2018) ..... 477

Affidavit of Emily Freedman (Oct. 24, 2018) ...... 484

Excerpts from Affidavit of Dr. Hermann Habermann (Oct. 26, 2018) ...... 489

Affidavit of Katherine Harvell Haney (Oct. 24, 2018) ...... 497

Affidavit of Jason Harmon (Oct. 23, 2018) ...... 502

Declaration of Samer Khalaf (Oct. 26, 2018) ...... 511

Affidavit of Christine Pierce (Nov. 6, 2018) ...... 527

Affidavit of Elizabeth Plum (Oct. 26, 2018) ...... 547

Affidavit of Jacqueline Tiema-Massie (Oct. 23, 2018) ...... 534

Volume 3 of 4

Declaration of Daniel Altschuler (Oct. 26, 2018) ...... 558 vi

PAGE

Supplemental Declaration of Daniel Altschuler (Nov. 2, 2018) ...... 572

Affidavit of Todd A. Breitbart (Oct. 26, 2018) ...... 583

Excerpts from Amended Declaration of Dr. John Thompson (Nov. 11, 2018) ...... 592

Excerpts from U.S. Census Bureau Statistical Quality Standards ...... 604

Memorandum from Lisa Blumerman to the Record (Apr. 29, 2016) ...... 634

Excerpts from Defendants’ Objections and Responses to Plaintiffs’ Requests for Admission to Defendant U.S. Department of Commerce ...... 650

Excerpts from Office of Management and Budget Standards and Guidelines for Statistical Surveys (Sept. 2006) ...... 654

Addendum to Statistical Policy Directive No. 2 ..... 677

Committee on National Statistics Letter Report on the 2020 Census (Aug. 7, 2018) ...... 697

Excerpts from Status Conference Transcript (July 3, 2018) ...... 747

Excerpts from Trial Transcript (Nov. 5, 2018–Nov. 15, 2018) ...... 755

Excerpts from 2010 Census Match Study (Nov. 16, 2012) ...... 953 vii

PAGE

Testimony of Wilbur Ross before the House Committee on Ways and Means (Mar. 22, 2018) ...... 956

Defendant’s Post-Trial Proposed Findings of Fact and Conclusions of Law Regarding Plaintiffs’ Claims ...... 960

Volume 4 of 4

Disclosure Avoidance Officer Exempt Data Product Release Form (Jan. 22, 2018) (A.R. 10735–10736) ...... 965

Excerpts from J. David Brown, et al., Understanding the Quality of Alternative Citizenship Data Sources for the 2020 Census (Aug. 6, 2018) ...... 967

Excerpts from Deposition of John Gore (Oct. 26, 2018) ...... 1021

Excerpts from Deposition of Sahra Park-Su (Oct. 25, 2018) ...... 1116

Excerpts from Deposition of Dr. John Abowd (Aug. 29, 2018) ...... 1119

Declaration of Dr. Andrew Reamer (Oct. 23, 2018) ...... 1136

Excerpts from Declaration of Dr. Christopher Warshaw (Oct. 26, 2018) ...... 1164

viii

PAGE

Census Bureau Website Page: “American Community Survey When to Use 1-Year, 3-Year, or 5-Year Estimates” ...... 1197

U.S. Census Bureau Proposed Information Collection, 2020 Census Federal Register Notice Vol. 83, No. 111 (June 8, 2018) ...... 1200

2000 Long-Form Decennial Census (A.R. 2–39) ... 1211

2000 Short-Form Decennial Census (A.R. 42–46) ...... 1249

2010 Decennial Census (A.R. 48–52) ...... 1254

2010 American Community Survey (ACS) (A.R. 509–523) ...... 1259

Excerpts from Deposition of Earl Comstock ...... 1274

Excerpts from 2020 Census Program Management Review (A.R. 5253–5275)...... 1295 1601

[COVER] Subjects Planned for the 2020 Census and American Community Survey ______

Federal Legislative and Program Uses

Issued March 2017 Revised

[IMAGE]

United StatesTM Census Bureau U.S. Department of Commerce Economics and Statistics Administration U.S. CENSUS BUREAU census.gov

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 1612

*** Age Age asked since 1790.

AGE AND DATE OF BIRTH QUESTIONS ARE USED TO UNDERSTAND THE SIZE AND CHARACTERISTICS OF DIFFERENT AGE GROUPS AND TO PRESENT OTHER DATA BY AGE. Age data are used in planning and funding government programs that provide funds or services for specific age groups, such as children, working-age adults, women of childbearing age, or the older population. These statistics are also used to enforce laws, regulations, and policies against age discrimination in government programs and in society.

AGE DATA HELP COMMUNITIES:

Provide Assistance to Older Americans Knowing how many people in a community are aged 60 and older helps local officials provide programs and services that enable older adults to remain living safely in their homes and communities (Older Americans Act). Age data are also used in programs that provide services and assistance to seniors, such as financial assistance with utilities (Low Income Home Energy Assistance Program).

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 1623

Provide Assistance to Children and Families Knowing the numbers and ages of children in families in combination with other information, such as household income, health insurance status, and poverty status, can help communities enroll eligible families in programs designed to assist them. For example, age data are used in targeted efforts to enroll eligible people in Medicaid and the Children’s Health Insurance Program.

Educate Children and Adults Knowing how many children and adults depend on services through schools helps school districts make long-term building, staffing, and funding decisions. Age in combination with other information, such as disability status, language spoken at home, and poverty status, assists schools in understanding the needs of their students and qualifying for grants that help fund programs for those students (Elementary and Secondary Education Act of 1965).

Ensure Equal Opportunity Knowing the ages of people in the community in combination with information about housing, employment, and education, helps government and communities enforce laws, regulations, and policies against discrimination based on age. For example, age information is used to analyze the employment status of workers by age (Age Discrimination in Employment Act).

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 1634 and 1490a 7 CFR 3550.10 6337(b)(1)(A) 7801(20) Patient Protection and Affordable Care Act, Public Law 111-148, § 10334; 42 USC 300kk 28 CFR Part 55 U.S. Department of EducationU.S. Department of Education 20 USC §§ 6333, 6334(a)(1), 6335(a), and U.S. Department of Health and Human Services, Administration 220 USC §§ 6821, 6824, 7011(5), and 42 USC 300kk Human Services, Center for U.S. Department of Agriculture 42 USC §§ 1472, 1474, 1485, 1486, 1490, for Community Living U.S. Department of Health and Medicare and Medicaid Services U.S. Department of Health andHuman Services, Office of the Assistant Secretary for Planning and Evaluation 42 USC § 1397ii(b)(2)(A)-(C) U.S. Department of Housing and Urban Development 12 USC § 1701q; 24 CFR part 891 U.S. Department of Justice,Civil Rights Division Voting Rights Act of 1965, Public Law 89- 110, as amended, § 203; 52 USC 10503; Selected Statutory Uses of Age Data

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 82 CU•APPENDIX • ACLU • 78228

01 (DOCUMENTS 1THRU6) U.S. Department of Justice, Title VII of the Civil Rights Act of 1964, Civil Rights Division Public Law 88-352; 42 USC § 2000e-2 U.S. Department of Labor Older Americans Act Amendments of 2000, Public Law 109-365, 42 USC § 3056e; 20 CFR 641.140, 641.360, and 641.365 U.S. Department of Labor 29 USC §§ 49f(a)(3)(D), 49g(d), and 49l-2(a)15 U.S. Department of Transportation Fixing America’s Surface Transportation Act, Public Law 114-94; 49 USC § 5303(c),

(e), (h), (i), (j),(k), and (n) 164 5 U.S. Department of Veterans Affairs 38 USC § 8104(b)(2) U.S. Environmental Protection Federal Water Pollution Control Act Agency (Clean Water Act), Public Law 92-500, 33 USC § 1254(a)(1)-(2), (b)(2), (b)(6), (b)(7), (n)(1), and (o)(1) U.S. Equal Employment Opportunity Age Discrimination in Employment Act of Commission, Office of General 1967, Public Law 90-202, 29 USC § 623(a)-(d) Counsel and 633a; 29 CFR 1625.7(d); Hazelwood School Dist. v. United States, 433 U.S. 299 (1977) U.S. Social Security Administration The Social Security Act, Public Law 74-271, as amended, 42 USC § 401(c) AL 3/21/19 1656

Gender Gender asked since 1790.

A QUESTION ABOUT THE GENDER OF EACH PERSON IS USED TO CREATE STATISTICS ABOUT MALES AND FEMALES AND TO PRESENT OTHER DATA, SUCH AS OCCUPATION, BY GENDER. Gender data are used in planning and funding government programs and in evaluating other government programs and policies to ensure they fairly and equitably serve the needs of males and females. These statistics are also used to enforce laws, regulations, and policies against discrimination in government programs and in society.

GENDER DATA HELP COMMUNITIES:

Ensure Equal Opportunity Knowing the gender of people in the community in combination with information about housing, voting, language, employment, and education, helps government and communities enforce laws, regulations, and policies against discrimination on the basis of gender. For example, gender data are used to enforce laws against discrimination based on gender in education programs and activities receiving federal financial assistance (Title IX of the Education Amendments of 1972).

Understand Changes Knowing whether people of different genders have the same opportunities in education, employment,

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 1667

voting, home ownership, and many other areas is of interest to researchers, advocacy groups, and policymakers. For example, the National Science Foundation uses gender data to provide information on women in the science and engineering workforce, and several agencies use gender data to investigate whether women, including women who are military veterans, have similar employment opportunities as men.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 1678 USC § 8623(a)(2) and (4), 8629(a)(1)-(3), Human Services, Administration and (6), § 8629(b) U.S. Department of Health and for Children and Families 42 U.S. Department of Health and Human Services, Administration for Community Living 42 USC 300kk U.S. Department of Health andHuman Services, Agency forHealthcare Research and Quality 42 USC §§ 299a(a)(3),(6),(8), 299b-2(a)(1), U.S. Department of Health and Human Services, Center for and 299(c)(1)(B) Medicare and Medicaid Services Patient Protection and Affordable Care Act, U.S. Department of Health and Human Services, Health Resources 42 USC 300kk Public Law 111-148, § 10334; and Services Administration, 42 USC § 254e; CFR 5.2 Bureau of Clinician Recruitment and Service Selected Statutory Uses of Gender Data

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 82 CU•APPENDIX • ACLU • 78228 01 (DOCUMENTS 1THRU6)

U.S. Department of Health and 42 USC § 1397ii (b)(2)(A)-(C) Human Services, Office of the Assistant Secretary for Planning and Evaluation U.S. Department of Housing and Fair Housing Act, Public Law 90-284, Urban Development 42 USC 3600-3620, 42 USC 3608(e) U.S. Department of Justice, Title VII of the Civil Rights Act of 1964, 168

Civil Rights Division Public Law 88-352, 42 USC § 2000e(2)(k); 9 Wards Cove Packing Co. v. Atonio; 490 U.S. 642 (1989) U.S. Department of Justice, Title IX of the Education Amendments of Civil Rights Division 1972, 20 USC § 1701 et seq. U.S. Department of Transportation Fixing America’s Surface Transportation Act, Public Law 114- 94; 49 USC § 5303(c), (e), (h), (i), (j),(k), and (n) U.S. Department of Veterans Affairs 38 USC § 546 AL 3/21/19 82 CU•APPENDIX • ACLU • 78228 01 (DOCUMENTS 1THRU6)

U.S. Environmental Protection Federal Water Pollution Control Act Agency (Clean Water Act), Public Law 92-500, 33 USC § 1254(a)(1)-(2), (b)(2), (b)(6), (b)(7), (n)(1), and (o)(1) U.S. Equal Employment Civil Rights Act of 1964, Public Law 88-352; Opportunity Commission, 42 USC § 2000e-2(k)(1)(A); Hazelwood 169 Office of General Counsel School Dist. v. United States, 433 U.S. 299, 10 307-308 (1977) U.S. Equal Employment Opportunity Civil Rights Act of 1964, Public Law 88-352; Commission, Office of Research, 42 USC § 2000e-2(k)(1)(A); Hazelwood Information, and Planning School Dist. v. United States, 433 U.S. 299, 307-308 (1977) U.S. Social Security Administration The Social Security Act, Public Law 74-271, as amended, 42 USC § 401(c) AL 3/21/19 17011

Race/Ethnicity Race asked since 1790, ethnicity asked since 1970.

QUESTIONS ABOUT A PERSON’S RACE OR ETHNICITY ARE USED TO CREATE DATA ABOUT RACE AND ETHNIC GROUPS. These data are required for federal and state programs and are critical factors in the basic research behind numerous policies, particularly for civil rights. Race and ethnicity data are used in planning and funding government programs that provide funds or services for specific groups. These data are also used to evaluate government programs and policies to ensure they fairly and equitably serve the needs of all racial and ethnic groups and to monitor compliance with antidiscrimination laws, regulations, and policies. States also use these data to meet legislative redistricting requirements. The U.S. Census Bureau collects race and ethnicity data in accordance with the 1997 Office of Management and Budget standards on race and ethnicity. The categories on race and ethnicity are based on self-identification and generally reflect a social definition of race and ethnicity. The categories are not an attempt to define race and ethnicity biologically, anthropologically, or genetically.

RACE AND ETHNICITY DATA HELP COMMUNITIES:

Ensure Equal Opportunity Knowing the races and ethnicities of community members in combination with information about

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 17112

housing, voting, language, employment, and education, helps government and communities enforce antidiscrimination laws, regulations, and policies. For example, race and ethnicity data are used in the following ways: • Establish and evaluate the guidelines for federal affirmative action plans under the Federal Equal Opportunity Recruitment Program. • Monitor compliance with the Voting Rights Act and enforce bilingual requirements. • Monitor and enforce equal employment opportunities under the Civil Rights Act of 1964. • Identify segments of the population who may not be getting needed medical services under the Public Health Service Act. • Allocate funds to school districts for bilingual services under the Bilingual Education Act.

Understand Changes Knowing if people of different races and ethnicities have the same opportunities in education, employment, voting, home ownership, and many other areas is of interest to researchers, advocacy groups, and policymakers. The National Science Foundation uses data on race and ethnicity to provide information on people of different racial and ethnic backgrounds in the science and engineering workforce. Several federal agencies use race and ethnicity data to investigate whether housing or transportation improvements have unintended consequences for specific race

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 17213

and ethnic groups. Data on race and ethnicity are used with age and language data to address language and cultural diversity needs in health care plans for the older population.

Administer Programs for Specific Groups Knowing how many people are eligible to participate in certain programs helps communities, including tribal governments, ensure that programs are operating as intended. For example, the Indian Housing Block Grant program, Indian Community Development Block Grant program, and Indian Health Service all depend on accurate estimates of American Indians and Alaska Natives. Data for the American Indian and Alaska Native population come from the questions about a person’s race or ethnicity.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 17314 Community Services Block Grant Act, Public Snyder Act, Nov. 2, 1921, c. 115, 25 USC § 13; ment Act, Public Law 94-43; 25 USC § 1602 Human Services, Administration Law 105-285, 42 USC §§ 9902(2), 9903, and Human Services, Indian HealthService Transfer Act, Aug. 5, 1954, c. 658, § 2, 42 USC § 2001(a); Indian Healthcare Improve- U.S. Department of Commerce, Bureau of the Census U.S. Department of Commerce, 13 USC § 141(c) Bureau of the Census U.S. Department of Health and 52 USC § 10503 for Children and FamiliesU.S. Department of Health and Human Services, Administration for Community Living Older Americans Act of 1965, Public Law 89-73, 42 USC § 3018 U.S. Department of Health and Human Services, Administration 9908(b)(1)(A), (b)(11), and (c)(1)(A)(i) for Community Living 42 USC 300kk U.S. Department of Health andHuman Services, Center for Medicare and Medicaid Services Patient Protection and Affordable Care Act, U.S. Department of Health and 42 USC § 300kk Public Law 111- 148, § 10334; Selected Statutory Uses of Race/Ethnicity Data

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 17415 91.205(a)-(c) 24 CFR 1000.324-1000.330 (n)(1), and (o)(1) Civil Rights Act of 1964, Public Law 88-352, 307-308 (1977) 12705(b)(1)-(3); 24 CFR Part 91, Housing and Community Development Act of 104-330, as amended, 25 USC § 4152(b); USC § 1254(a)(1)-2), (b)(2), (b)(6), (b)(7), U.S. Department of Housing and McKinney-Vento Homeless Assistance Act, U.S. Department of Justice, Title VII of the Civil Rights Act 1964, U.S. Equal Employment Opportunity Commission, Office of Research,Information, and Planning 42 USC § 2000e-2(k)(1)(A); Hazelwood School Dist. v. United States, 433 U.S. 299, U.S. Department of Housing and Urban Development Cranston-Gonzalez National Affordable Urban DevelopmentU.S. Department of Housing and Urban DevelopmentU.S. Department of Housing and Housing Act, Public Law 101-625, 42 USC Urban Development Native American Housing Assistance and 42 USC 11371-11376; 24 CFR Part 91 Civil Rights DivisionU.S. Department of Justice, 1974, 42 USC 5306(a)(1); 24 CFR §1003.101 Civil Rights DivisionU.S. Environmental Protection Self-Determination Act of 1996, Public Law Agency Voting Rights Act of 1965, § 203, Federal Water Pollution Control Act 42 USC § 2000e-2 52 USC § 10503; 28 CFR Part 55 (Clean Water Act), Public Law 92-500, 33

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 17516

Relationship Relationship asked since 1880.

A QUESTION ABOUT THE RELATIONSHIP OF EACH PERSON IN A HOUSEHOLD TO ONE CENTRAL PERSON IS USED TO CREATE ESTIMATES ABOUT FAMILIES, HOUSE- HOLDS, AND OTHER GROUPS, AND TO PRESENT OTHER DATA AT A HOUSEHOLD LEVEL. Relationship data are used in planning and funding government programs that provide funds or services for families, people living or raising children alone, grandparents living with grandchildren, or other households that qualify for additional assistance.

RELATIONSHIP DATA HELP COMMUNITIES:

Provide Adequate Housing Knowing about the different types of households in a community (single people, couples, families, roommates, etc.) helps communities understand whether available housing meets the needs of residents. Information about the relationships among people in a household, in combination with housing costs and the combined income of all people in a household, helps communities understand whether housing is affordable for residents. When housing is not sufficient or not affordable, relationship data can help communities enroll eligible households in programs designed to assist them, and can help communities qualify for grants from the Community Development Block Grant,

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 17617

HOME Investment Partnership Program, Emergency Solutions Grant, Housing Opportunities for Persons With AIDS, and other programs.

Provide Assistance to Families Knowing more about families, such as the ages of children, household income, health insurance status, and poverty status, can help communities enroll eligible families in programs designed to assist them, such as Head Start and the Children’s Health Insurance Program, and can help communities qualify for grants to fund these programs. Relationship data are also used to ensure that programs like Temporary Assistance for Needy Families are making a difference for families.

Understand Changing Households Information about living arrangements and how they are changing, including whether older residents are staying in their homes as they age, whether young people are living with parents or moving in with roommates, and which kinds of households include young children, can help communities plan future programs and services for residents. For example, the Social Security Administration estimates future program needs based on the current relationships of working people.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 17718 (b), and 8622 (11) USC §§ 8629 (a) (1)-(3) and (5)-(6), § 6861, 6864; 10 CFR 440.10 Act, Public Law 114-94; 49 USC § 5303(c), (e), (h), (i), (j),(k), and (n) 6337(b)(1)(A) 8629 Public Law 94-385, as amended, 42 USC U.S. Department of Transportation Fixing America’s Surface Transportation U.S. Department of EducationU.S. Department of Health and Human Services, Administration 20 USC §§ 6333, 6334(a)(1), 6335(a), 42 U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 42 USC § 242k(b), (h), and (l) National Center for Health Statistics U.S. Department of Energy Energy Conservation and Production Act, for Children and Families U.S. Department of Health and Human Services, Administration for Children and Families 13 USC § 141 note U.S. Department of Health and Human Services, Administrationfor Community Living Developmental Disabilities Assistance and Bill of Rights Act 2000, Public Law 106- 402, § 124(c)(5); 42 USC 15024 Selected Statutory Uses of Relationship Data

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 82 CU•APPENDIX • ACLU • 78228

U.S. Department of Health and Patient Protection and Affordable Care Act,

01 (DOCUMENTS 1THRU6) Human Services, Center for Public Law 111-148, §10334; 42 USC 300kk Medicare and Medicaid Services U.S. Department of Health and 42 USC § 1397ii (b)(2)(A)-(C) Human Services, Office of the Assistant Secretary for Planning and Evaluation U.S. Department of Housing and McKinney-Vento Homeless Assistance Act, Urban Development 42 USC 11371- 11376; 24 CFR Part 91 U.S. Department of Housing and Housing and Community Development Act Urban Development of 1974, as amended, Public Law 93-383, 42 178 USC 5301, 5302, and 5305; 24 CFR 91.205(a)- 19 (c ), 91.305(a)-(c), 570.208(a)(1), 570.483(b)(1), 570.704(a)-(c), 570.707(a)-(c), and 570.901 U.S. Department of Housing and Cranston-Gonzalez National Affordable Urban Development Housing Act, Public Law 101-625, 42 USC 12705(b)(1)-(3); 24 CFR Part 91; 24 CFR 91.205(a)-(c) U.S. Social Security Administration The Social Security Act, Public Law 74-271, as amended, 42 USC § 401(c) U.S. Department of Veterans Affairs Veterans Benefits Improvement Act of 2008, Public Law 110-389, Title III—Labor and Education Matters, Subtitle C—Vocational Rehabilitation Matters, § 334, 38 USC § 3122 AL 3/21/19 17920

Tenure (Owner/Renter) Tenure asked since 1890.

A QUESTION ABOUT WHETHER A HOME IS OWNED OR RENTED IS USED TO CREATE DATA ABOUT TENURE, RENTERS, AND HOME OWNERSHIP. Tenure is the most basic characteristic to asses housing inventory. Tenure data are used in government programs that analyze whether adequate housing is affordable for residents. Tenure data are also used to provide and fund housing assistance programs. These statistics are also used to enforce laws, regulations, and policies against discrimination in private-market housing, government programs, and in society.

TENURE DATA HELP COMMUNITIES:

Provide Adequate Housing Knowing the different types of households in a community (single people, couples, families, roommates, etc.) and rates of home rental and ownership helps communities understand whether available housing meets the needs of residents. Data about owners and renters, in combination with housing costs and the combined income of all people in a household, help communities understand whether housing is affordable for residents. When housing is not sufficient or affordable, data about owners and renters can help communities enroll eligible households in programs designed to assist them, and can help communities qualify for grants from the Community Development Block

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 18021

Grant, HOME Investment Partnership Program, Emergency Solutions Grant, Housing Opportunities for Persons With AIDS, and other programs.

Plan Community Development Knowing how the balance of rented homes, mort- gaged homes, and homes owned free and clear changes over time can help communities understand changes in local housing markets; identify opportunities to improve tax, assistance, and zoning policies; and to reduce tax revenue losses from vacant or abandoned properties. Tenure is also used in formulas that communities use to determine housing assistance funding (Fair Market Rents).

Ensure Equal Opportunity Knowing the characteristics of people who rent and people who own homes in the community, such as age, gender, race, Hispanic origin, disability, helps government and communities enforce laws, such as the 1968 Fair Housing Act, designed to eliminate discrimination in housing.

Understand Changing Households Knowing whether older residents are staying in homes as they age or moving into rented homes; and whether young people are staying with parents, renting with roommates, or buying homes, can help governments and communities distribute funds appropriately between home ownership and rental housing programs and services for residents.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 18122 1490a, 1490I, 1490m, 1490p-2, 1490r; 7 CFR 1940.563-564, 1940.575, 3560.11, and 3560.152(a)(2) 42 USC § 1439 (d)(1)(A)(i); 24 CFR 791.402 1437f(c)(1); 24 CFR 888.113; 982.401 12705(b)(1)-(3); 24 CFR Part 91, 91.205(a)-(c) 24 CFR § 8.23(a) U.S. Department of Agriculture 42 USC §§ 1472, 1474, 1485, 1486, 1490, Urban DevelopmentU.S. Department of Housing and Urban Development Housing and Community Development Act 42 USC § 11371-11376; 24 CFR Part 91 of 1974, Public Law 93-383, as amended, U.S. Department of Housing and McKinney-Vento Homeless Assistance Act, U.S. Department of Housing and Urban Development United States Housing Act of 1937, Public U.S. Department of Housing andUrban Development Cranston-Gonzalez National Affordable U.S. Department of Housing and Law 93-383, as amended, 42 USC § Urban Development Rehabilitation Act of 1973, § 504, Public Housing Act, Public Law 101-625, 42 USC Law 93-112, 29 USC 794; 24 CFR § 8.22(b); Selected Statutory Uses of Tenure Data

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 82 CU•APPENDIX • ACLU • 78228

01 (DOCUMENTS 1THRU6) U.S. Department of Housing and 12 USC § 4568 Urban Development U.S. Department of Housing and 12 USC § 1701q; 24 CFR part 891 Urban Development U.S. Department of Housing and Tax Reform Act of 1986, Public Law 99-514, Urban Development 26 USC § 42(d)(5)(B)(ii)(I), (iii)(I), (iv), and(g); 15 U.S.C § 631 U.S. Department of Justice, Voting Rights Act of 1965; 52 USC § 10301; 182 Civil Rights Division 28 CFR Part 51; LULAC v. Perry, 548 U.S. 23 399 (2006); Johnson v. DeGrandy, 512 U.S. 997 (1994); Thornburg V. Gingles, 478 U.S. 30 (1986) U.S. Department of Transportation 49 USC § 5303; 49 CFR Part 613 U.S. Department of Transportation Fixing America’s Surface Transportation Act, Public Law 114-94; 49 USC § 5303(c), (e), (h), (i), (j),(k), and (n) U.S. Department of Transportation 49 USC §§ 6302(b)(3)(B), 6302(c), 6304(a), 6309(a) AL 3/21/19 18324

PAGE 248

Place of Birth, Citizenship, and Year of Entry Place of birth asked since 1850, citizenship asked since 1820,1 year of entry asked since 1890.2

QUESTIONS ABOUT A PERSON'S PLACE OF BIRTH, CITIZENSHIP, AND YEAR OF ENTRY INTO THE UNITED STATES ARE USED TO CREATE DATA ABOUT CITIZENS, NON- CITIZENS, AND THE FOREIGN-BORN POPULATION. These statistics are essential for agencies and policymakers setting and evaluating immigration policies and laws, seeking to understand the experience of different immigrant groups, and enforcing laws, policies, and regulations against discrimination based on national origin. These statistics are also used to tailor services to accommodate cultural differences.

PLACE OF BIRTH, CITIZENSHIP, AND YEAR OF ENTRY DATA HELP COMMUNITIES:

Ensure Equal Opportunity Knowing how many people in the community are born in other countries in combination with information about housing, voting, language, employment, and education, helps the government and communities to enforce laws, regulations, and

1 Citizenship asked 1820-1830, 1870, and 1890 to present. 2 Year of entry asked 1890-1930, and 1970 to present.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 18425

policies against discrimination based on national origin. For example, these data are used to support the enforcement responsibilities under the Voting Rights Act to investigate differences in voter participation rates and to enforce other laws and policies regarding bilingual requirements.

Educate Children Knowing how many foreign-born children depend on services through schools helps school districts make staffing and funding decisions. Place of birth, citizenship, and year of entry statistics in combination with other information, such as language spoken at home, help schools understand the needs of their students and qualify for grants that help fund programs for those students (Elementary and Secondary Education Act of 1965).

Understand Changes Knowing whether people of different races or countries of birth have the same opportunities in education, employment, voting, home ownership, and many other areas is of interest to researchers, advocacy groups, and policymakers. These data may also help communities with large refugee populations that qualify for financial assistance (Immigration Nationality Act).

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 18526

PX-019

NOT RELEVANT

From: Kris Kobach [mailto [REDACTED] Sent: Monday, July 24, 2017 2:43 PM To: Teramoto, Wendy (Federal) Cc: Alexander, Brooke (Federal) [REDACTED]; Hernandez, Israel (Federal) [REDACTED] Subject: Re: Follow up on our phone call Yes. Sent from my iPhone On Jul 24, 2017, at 1:39 PM, Teramoto, Wendy (Federal) [REDACTED] wrote: Kris – can you do a call with the Secretary and Izzy tomorrow at 11 am? Thanks. Wendy From: Kris Kobach [mailto: [REDACTED] Sent: Monday, July 24, 2017 12:02 PM To: Teramoto, Wendy (Federal) [REDACTED] Subject: Re: Follow up on our phone call That works for me. What number should I call? Or would you like to call me? On Mon, Jul 24, 2017 at 9:12 AM, Teramoto, Wendy (Federal) [REDACTED] wrote: We can speak today at 230. Please let me know if that works. W Sent from my iPhone

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 18627

On Jul 21, 2017, at 4:34 PM, Kris Kobach [REDACTED] wrote: Wendy, Nice meeting you on the phone this afternoon. Below is the email that I sent to Secretary Ross. He and I had spoken briefly on the phone about this issue, at the direction of Steve Bannon, a few months earlier. Let me know what time would work for you on Monday, if you would like to schedule a short call. The issue is pretty straight forward, and the text of the question to be added is in the email below. Thanks. Kris Kobach ------Forwarded message ------From: Kris Kobach [REDACTED] Date: Fri, Jul 14, 2017 at 9:12 AM Subject: Follow up on our phone call To: [REDACTED] Secretary Ross, Kansas Secretary of State Kris Kobach here. I’m following up on our telephone discussion from a few months ago. As you may recall, we talked about the fact that the US census does not currently ask respondents their citizenship. This lack of information impairs the federal government’s ability to do a number of things accurately. It also leads to the problem that aliens who do not actually “reside” in the United States are still counted for congressional apportionment purposes.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 18728

It is essential that one simple question be added to the upcoming 2020 census. That question already appears on the American Community Survey that is conducted by the Census Bureau (question #8). A slight variation of that question needs to be added to the census. It should read as follows: Is this person a citizen of the United States? Ⅺ Yes, born in the United States Ⅺ Yes, born in Puerto Rico, Guam, the U.S. Virgin Islands, or Northern Marianas Ⅺ Yes, born abroad of U.S. citizen parent or parents Ⅺ Yes, U.S. citizen by naturalization – Print year of naturalization ---____ Ⅺ No, not a U.S. citizen – this person is a lawful permanent resident (green card holder) Ⅺ No, not a U.S. citizen – this person citizen of another country who is not a green card holder (for example holds a temporary visa or falls into another category of non-citizens) Please let me know if there is any assistance that I can provide to accomplish the addition of this question. You may reach me at this email address or on my cell phone at [REDACTED] Yours, Kris Kobach

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 18829

PAGES 773 TO 775 The Leadership Conference on Civil and Human Rights

[LETTERHEAD] [STAMP] 2018 FEB 22 PM 2:39 O S EXECUTIVE SECRETARIAT [HANDWRITTEN] 18-059793

January 4, 2018 Honorable Wilbur L. Ross Secretary of Commerce U.S. Department of Commerce 1401 Constitution Ave NW Washington, DC 20230 Dear Secretary Ross: I write as the president and CEO of The Leader- ship Conference on Civil and Human Rights, a coalition charged by its diverse membership of more than 200 national organizations to promote and protect the civil and human rights of all persons in the United States. In this capacity and on behalf of this broad coalition, I urge you to reject the request in the December 12, 2017 letter from the Department of Justice to Acting Census Director Ron Jarmin, to add a new citizenship question on the 2020 Census. As you well know, adding a new and untested question to the 2020 Census would disrupt preparations at a pivotal point in the decade, undermine years of costly, painstaking research and testing, and increase

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 18930

census costs significantly at a time when Congress has directed a less expensive enumeration. All of these factors would threaten a fair and accurate decennial census. We appreciate the commitment to a full, fair, and accurate census that you and your senior staff have recently expressed. The Leadership Conference views a fair and accurate census, and the collection of useful, objective data about our nation’s people, housing, economy, and communities generally, to be among the most important civil rights issues of our day. However, as discussed below, the Justice Department’s ill-advised proposal poses a significant threat to our shared goal. First, as you noted during the House Committee on Oversight and Government Reform’s October 12, 2017 hearing on the 2020 Census (where we both testified), requiring a new topic this late in the preparations for the census is irresponsible because robust testing for new questions in a contemporary, census-like environment is essential. This is especially true given the chilling effect of adding a citizenship question to the form. Census preparations are already behind schedule, the final dress rehearsal will kick off in a month, and there simply is no time left to redesign the census form and rigorously test the proposed additional question. As we know from extensive research and testing in the survey field, even small changes in question order and wording can significantly affect both the rate and accuracy of responses. Yet the Census Bureau has neither the time nor the resources to evaluate the consequences of such a major change in the questionnaire.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19031

Second, as I know from my prior experience as the chief government enforcer of the Voting Rights Act, the Justice Department has never needed to add this new question to the decennial census to enforce the Voting Rights Act before, so there is no reason it would need to do so now. Contrary to the Justice Department’s letter, the Census Bureau has ---not included a citizenship question on the modem census “short form,” sent to every household. In fact, no such question has appeared on the census “short form” since enactment of the Voting Rights Act in 1965. Estimates of the citizen voting-age population derived from the ongoing American Community Survey, and the so-called census “long” or sample form before that, have been and continue to be suitable for purposes of civil rights and Voting Rights Act enforcement. Whether utilizing such data for Section 2 enforcement actions, Section 203 determinations, or other voting rights enforcement efforts, courts and the Justice Department have accepted census data as currently collected since enactment of the Voting Rights Act. Civil rights groups, likewise, have never asserted a need for a “100 percent” census citizenship question in order to effectively represent and ensure voting rights for minority communities. Given these plain facts, the entire justification for the request should be viewed skeptically as an attempt to throw a wrench into final planning and preparations for an enumeration that already faces enormous challenges, including inadequate and delayed funding, cyber-security risks, and a climate of fear fanned by anti- immigrant rhetoric.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19132

Third, this new proposed question on the 2020 Census is unnecessarily intrusive and will raise concerns in all households – native and foreign born, citizens and noncitizens – about the confidentiality of information provided to the government and how government authorities might use that information. Asking every household and every person in the country about their citizenship status in the current political environment – when there is no legal basis or need for doing so – will no doubt give many people pause about participating in the census altogether. In fact, new Census Bureau research already is raising alarm bells about the growing reluctance of immigrant households to participate fully and honestly in any Census Bureau surveys, due to their fear about how their responses will be used by government agencies. Adding this new question would also result in taxpayers spending significantly more for a government undertaking that we know in advance would not be successful. Your recent oral testimony before Congress acknowledged that the Census Bureau will need billions of dollars more than originally estimated to conduct a modem, inclusive census. The Justice Department’s proposal to add a new citizenship question would increase census costs even further while decreasing accuracy, because self-response rates are certain to plummet, which in turn will require additional, costly door- to-door visits that still may not spur cooperation or result in accurate responses. Finally, this request – coming almost a year after the Census Bureau has finalized topics for the 2020 Census, as required by law – risks jeopardizing

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19233

the accuracy of the 2020 Census in every state and every community by deterring many people from responding, making the data collected in this crucial once-a-decade operation less accurate and useful for all of us. As four former Census Directors, who served in both Republican and Democratic administrations, wrote in an amici curiae brief in the Supreme Court case Evenwel v. Abbott, asking about citizenship status in the decennial census ‘‘would likely exacerbate privacy concerns and lead to inaccurate responses from non-citizens worried about a government record of the immigration status... The sum effect would be bad Census data.” I know you appreciate that the stakes of a fair and accurate census are high and everyone – from Congress to governors, mayors, and school board officials, to business owners and nonprofits serving the most vulnerable in our communities – will have to live with any flawed results for the next 10 years. The Leadership Conference and its member organizations look forward to working with you and your staff to ensure a cost-effective, secure, and above all, accurate and inclusive census in every one of the nation’s communities. If you have any questions, please contact me or Chris Harley, Census Campaign Director, at (202) 466-3311. Thank you for your consideration of our views. Sincerely,

/s/ Vanita Gupta

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19334

Cc: Acting Deputy Secretary and Under Secretary for Economic Affairs Karen Dunn Kelley Acting Census Director Ron Jarmin PAGES 798 TO 803 The Leadership Conference on Civil and Human Rights

[LETTERHEAD] [STAMP] 2018 JAN 12 AM 8:00 O S EXECUTIVE SECRETARIAT [HANDWRITTEN] 18-059175

January 10, 2018

Protect the Census: Oppose DOJ Request to Add a Citizenship Question to the 2020 Census

Dear Secretary Ross: On behalf of The Leadership Conference on Civil and Human Rights, a coalition charged by its diverse membership of more than 200 national organizations to promote and protect the civil and human rights of all persons in the United States, and the undersigned 167 organizations, we urge you to reject the Department of Justice’s untimely and unnecessary request for a new citizenship question on the 2020 Census, which would threaten a fair and accurate decennial census. Adding a new citizenship question to the 2020 Census would destroy any chance for an

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19435

accurate count, discard years of careful research, and increase costs significantly. You and your staff have made clear that you share our goal of a full, fair, and accurate census. A fair and accurate census, and the collection of useful, objective data about our nation’s people, housing, economy, and communities generally, are among the most significant civil rights issues facing the country today. Every census since the first enumeration in 1790 has included citizens and non-citizens alike. Adding a new question on citizenship to the 2020 Census undoubtedly would affect response rates, outreach, and advertising strategies, and other important elements of the nation’s largest, most complex peacetime activity, calling into question the results of many years of costly, painstaking research and testing. Adding a citizenship question to the 2020 Census would disrupt preparations at a pivotal point in the decade, undermining years of research and testing and increasing census costs significantly at a time when Congress has directed a less expensive enumeration. The Justice Department’s request would literally would add billions of dollars to the life-cycle cost of this census, without improving accuracy. Questionnaire design and testing began nearly eight years ago during the 2010 Census. Requiring this new topic this late in the decade would threaten the success of the 2020 Census because robust testing in a census-like environment is essential, given the probable chilling effect of adding these questions to the form. There simply is no time to redesign the census form, craft

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19536

scientifically sound questions to collect the information the Justice Department requests, and evaluate the impact of this new question on census participation and operations before the census starts, in any responsible way. Given the constitutional requirement to conduct the census in 2020, final planning and preparations for the census would be haphazard, at best, leaving the nation with a deeply flawed foundation for our democratic ideals, informed decision-making, and prudent allocation of precious taxpayer dollars. In addition, adding this question would jeopardize the accuracy of the 2020 Census in every state and every community by deterring many people from responding. The question is unnecessarily intrusive and will raise concerns in all households – native- and foreign-born, citizens and noncitizens – about the confidentiality of information provided to the government and how that information might be used. Moreover, there are many mixed status households in the United States, which include members who are both citizens and non-citizens with various legal statuses. Mixed- status and immigrant households will be especially fearful of providing information to the federal government in 2020, given the heightened climate of fear that anti-immigrant rhetoric and policies have created. In short, any effort to determine citizenship through the constitutionally required census would jeopardize the accuracy of the entire count, leaving public, private, and nonprofit decision-makers with bad information for all purposes, for the next 10 years. Further, such an effort is likely to shake public confidence in the

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19637

narrow (though vital) statistical objectives of the Census Bureau’s work, damaging ongoing data collection efforts well into the future. Finally, in addition to being untimely, the request is unnecessary. The Justice Department has never needed to add this new question to the decennial census to enforce the Voting Rights Act before, so there is no reason it would need to·do so now. Contrary to the Justice Department’s letter, the Census Bureau has ---not included a citizenship question on the modern census “short form,” sent to every household. In fact, no such question has appeared on the census “short form” since enactment of the Voting Rights Act in 1965. Estimates of the citizen voting-age population derived from the ongoing American Community Survey, and the so-called census “long” or sample form before that, have been and continue to be suitable for purposes of civil rights and Voting Rights Act enforcement. Whether utilizing such data for Section 2 enforcement actions, Section 203 determinations, or other voting rights enforcement efforts, courts and the Justice Department have accepted census data as currently collected since enactment of the Voting Rights Act. Given these plain facts, the entire justification for the request should be viewed skeptically as an attempt to throw a wrench into final planning and preparations for an enumeration that already faces enormous challenges, including inadequate and delayed funding, cyber-security risks, and a climate of fear fanned by anti-immigrant rhetoric. For these reasons, we urge you to reject the Justice Department’s request to add a citizenship question to the 2020 Census. If you have any

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19738

questions about these comments, please contact Leadership Conference Census Task Force Co- chairs Terry Ao Minnis, Asian Americans Advanc- ing Justice|AAJC, at 202-296-2300 x0127, or Arturo Vargas, NALEO Educational Fund, at 213- 747-7606, or Chris Harley, Census Campaign Director at 202-466-3311. Sincerely, The Leadership Conference on Civil and Human Rights 18MillionRising.org ACCESS ACLU Advancement Project California Alliance for Strong Families and Communities American Association of University Women (AAUW) American Educational Research Association American Federation of State, County and Municipal Employees (AFSCME) American Federation of Teachers American Library Association American Sociological Association American Statistical Association American-Arab Anti-Discrimination Committee APACEvotes Arab American Institute Arkansas Advocates for Children and Families Asian & Pacific Islander American Health Forum Asian American Legal Defense and Education Fund (AALDEF) Asian Americans Advancing Justice – AAJC Asian Americans Advancing Justice – Asian Law Caucus

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19839

Asian Americans Advancing Justice – Atlanta Asian and Pacific Islander American Vote Asian Counseling and Referral Service Asian Pacific Islander Americans for Civic Empowerment (APACE) Association of Population Centers Association of Public Data Users Autistic Self Advocacy Network Black Women’s Roundtable Black Youth Vote! Bread for the World Brennan Center for Justice California Calls Campaign Legal Center Casa de Esperanza: National Latin@ Network for Healthy Families and Communities Casa Latina Center for American Progress Center for Law and Social Policy (CLASP) CenterLink: The Community of LGBT Centers Children Now Children’s Advocacy Alliance Church World Service Claritas Coalition for Disability Health Equity Coalition on Human Needs Common Cause Congregation Beth Shalom Congregation of Our Lady of Charity of the Good Shepherd, US Provinces Consortium of Social Science Associations Council for Community and Economic Research Council of Professional Associations on Federal Statistics Council on American-Islamic Relations (CAIR) Council on American-Islamic Relations, California

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 19940

Defending Rights & Dissent Detention Watch Network Education Law Center–PA Emgage Foundation Empowering Pacific Islander Communities Equal Justice Society Equality California Faith in Public Life Family Equality Council Farmworker Justice FORGE, Inc. Franciscan Action Network Friends Committee on National Legislation Friends of the Earth US Government Accountability Project Government Information Watch Hispanic Federation Human Rights Campaign Human Rights Watch In the Public Interest Indivisible Insights Association Interfaith Worker Justice Irish Immigration Center of Philadelphia Irish International Immigrant Center Islamic Society of North America, Office for SInterfaith and Community Alliances Jacobs Institute of Women’s Health Japanese American Citizens League KIDS COUNT in Delaware Lambda Legal LatinoJustice PRLDEF League of United Latin American Citizens League of Women Voters of the United States Legal Aid at Work

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20041

Los Angeles LGBT Center Maine Children’s Alliance MALDEF Massachusetts Voter Table Mi Familia Vota Muslim Justice League NAACP NAACP Legal Defense & Educational Fund, Inc. NALEO Educational Fund National Advocacy Center of the Sisters of the Good Shepherd National Alliance for Partnerships in Equity (NAPE) National Association of Social Workers National CAPACD National Center for Lesbian Rights National Center for Transgender Equality National Coalition on Black Civic Participation National Council of Asian Pacific Americans (NCAPA) National Council of Jewish Women National Council on Independent Living National Disability Rights Network National Education Association National Employment Law Project National Health Law Program National Immigrant Justice Center National Immigration Law Center National Institute for Latino Policy (NiLP) National Justice for Our Neighbors National Korean American Service & Education Consortium (NAKASEC) National Latina Institute for Reproductive Health National Latina/o Psychological Association National LGBTQ Task Force Action Fund

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20142

National Low Income Housing Coalition National Network for Arab American Communities National Partnership for Women & Families National Youth Employment Coalition Natural Resources Defense Council NC Child NC Counts Coalition Neighborhood Action Coalition NETWORK Lobby for Catholic Social Justice New Mexico Voices for Children Northern California Grantmakers OCA – Asian Pacific American Advocates OneAmerica OpenTheGovemment People For the American Way PFLAG National PICO California Pierce County Labor Community Services Agency Planned Parenthood Federation of America PolicyLink Population Association of America Presente.org Prison Policy Initiative Public Citizen Research Advisory Services, Inc. Senior Executives Association Service Employees International Union (SEIU) Sikh Coalition SiX Action Society of American Archivists Southeast Michigan Census Council Southern California Grantmakers Southern Coalition for Social Justice Southern Echo Inc.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20243

State Voices Sunlight Foundation The Children’s Partnership The United Methodist Church – General Board of Church and Society The Voter Participation Center UnidosUS (formerly NCLR) Union for Reform Judaism Union of Concerned Scientists Voices Verdes Voices for Progress Voices for Vermont’s Children Voto Latino Wallingford Indivisible Washington Immigrant Solidarity Network Win/Win Network Women’s Voices Women Vote Action Fund

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20344

PX-116

January 26, 2018 The Honorable Wilbur L. Ross Secretary of Commerce U.S. Department of Commerce 14th St. and Constitution Avenue, NW Washington, DC 20230 Dear Secretary Ross: As former directors of the U.S. Census Bureau, serving under both Republican and Democratic administrations, we want to thank you for the care for the future of the Census Bureau you have displayed. We were, however, troubled to learn that the Department of Justice has recently asked the Bureau to add a new question on citizenship to the 2020 census. We are deeply concerned about the consequences of this possible action and hope that our objective observations provide a useful perspective before a final decision is made on this issue. We were encouraged by your testimony before the Census Bureau’s House and Senate authorizing committees last October. Your frank assessment of the status of 2020 Census preparations and your acknowledgment that the Bureau will need more resources to conduct an acceptably accurate enumeration were correct. Undoubtedly, your substantial private sector experience has informed your approach to the Bureau’s mission. Similarly, your experience as a census enumerator many years ago may have helped to shape your appreciation for the importance of the fair and accurate census our Constitution envisions, free from partisan influence

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20445

and guided by sound, well documented, scientifically driven decisions.1 There is a well-proven multi-year process to suggest and test new questions. We strongly believe that adding an untested question on citizenship status at this late point in the decennial planning process would put the accuracy of the enumeration and success of the census in all communities at grave risk. Your observation at the House Oversight and Government Reform Committee hearing on October 12, 2017 — that adding untested questions could reduce response rates — suggests that you have carefully considered respondent burden and other factors that contribute to public acceptance of censuses and surveys, as the window of opportunity to lock down census methods, operations, content, and infrastructure closes quickly. As you fully appreciate, planning a decennial census is an enormous challenge. Preparations for a census are complex, with each component related to and built upon previous research and tests. The critical ‘dress rehearsal’ for the 2020 Census (the 2018 End-to-End Census Test) is starting in Providence County, RI. Adding a citizenship question without a testing opportunity in a contemporary, census-like environment will invalidate the results

1 We think you will enjoy recalling that Kenneth Prewitt, a signer of this letter, was your crew leader in 1960. You were in the Harvard Business School, and he in the Harvard Divinity School; like you, he wanted to make some extra money over spring break. Ken was appointed a crew leader and recruited enumerators only from the HBS, knowing that they would carry out their duties efficiently. Indeed, they (you) did — your crew finished first in Boston, with the highest accuracy score in the city.

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20546

and lessons learned from the End-to-End test. Key assumptions underlying estimates of self-response, staffing needs, local office sites, and communication strategies will no longer be sound, calling into question cost projections that we know you have worked hard to validate and update. In addition, the Census Bureau would need to modify data capture and processing systems, language assistance and enumerator training materials, and web-based instructions for completing the census in the time remaining before the 2020 Census starts – all without the benefit of field testing. There are sound reasons that the Census Act requires the Bureau to submit to Congress the topics and actual questions it will include, three and two years, respectively, before Census Day. It is highly risky to ask untested questions in the context of the complete 2020 Census design. There is a great deal of evidence that even small changes in survey question order, wording, and instructions can have significant, and often unexpected, consequences for the rate, quality, and truthfulness of response. The effect of adding a citizenship question to the 2020 Census on data quality and census accuracy, therefore, is completely unknown. Also of import, overcoming unexpected obstacles that arise as 2020 Census operations unfold would add to the cost, without assurances that such efforts would yield a more accurate outcome. In summary, we believe that adding a citizenship question to the 2020 Census will considerably increase the risks to the 2020 enumeration. Because we share your goal of a “full, fair, and accurate census,” as the Constitution requires, we urge you to

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20647

consider a prudent course of action in response to the Justice Department’s untimely and potentially disruptive request. Please let us know if we can answer any questions or be of further assistance. Sincerely, Vincent P. Barabba (1973-1976; 1979-1981) Martha Farnsworth Riche (1994-1998) Kenneth Prewitt (1998-2001) Steven H. Murdock (2008-2009) Robert M. Groves (2009-2012) John Thompson (2013-2017)

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20748

State of Louisiana DEPARTMENT OF JUSTICE

[LETTERHEAD] [SEAL] Jeff Landry Attorney General [STAMP] 2018 FEB –8 PM 3:03 O S EXECUTIVE SECRETARIAT February 8,2018 The Honorable Wilbur Ross United States Department of Commerce 1401 Constitution Ave NW Washington, D.C. 20233-0001 Re: Request to Reinstate Citizenship Question on 2020 Census Questionnaire Dear Secretary Ross: As the Chief Legal Officer of the State of Louisiana, I write concerning a matter that is very important to the People of Louisiana. The use of the decennial Census to capture accurate data concerning citizenship, legal immigration, illegal immigration, and the distribution of the population is crucial to the functions of State government. It is also crucial in terms of insuring fair and equitable districting of the people’s representatives at the State and local level. This issue touches the heart of our democracy and the constitutional rights of every Louisiana citizen. As you are aware, from 1970 to 2000, the Census Bureau included a citizenship question on the “long form” questionnaire sent to nearly one in every six

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20849

households during each decennial census. After the 2000 Census, the Census Bureau ceased using the “long form” questionnaire. Instead, it replaced this form with the American Community Survey (ACS). The ACS is currently the Census Bureau’s only survey that collects information regarding citizenship and estimates citizen voting-age population. The ACS is sent to far fewer people – approximately one in every thirty-eight households each year, significantly changing the statistical integrity of the data. The ACS, while insufficient for a number of reasons, most importantly provides only estimates with a high margin of error. Because it is standard practice for States to apportion their legislative districts on the basis of the numbers provided by the Census Bureau’s decennial census, this issue is of critical importance. See Nat’l Conf. of State Legislatures Redistricting Law 2010 at 11 (2009). States frequently even find themselves mired in federal litigation lasting from one Census to the next. Currently, the decennial Census counts everyone regardless of the individual’s legal status and no longer provides any reliable citizenship data. Ultimately, this process dilutes the votes of all legally-eligible voters by improperly counting those ineligible to vote when determining the population for representative districts. Not only does this result in bolstering the representation of illegal immigrants and non-voting legal immigrants at the expense of the voting age citizenry, but also skews the data nationally and can result in some states losing representatives in Congress to other States. Furthermore, the Supreme Court has held that Section 2 of the Voting Rights Act prohibits “vote dilution” by state and local jurisdictions engaged in

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 20950

redistricting, which can occur when a racial group is improperly deprived of a single-member district in which it could form a majority. See Thornburg v. Gingles, 478 U.S. 30, 50 (1986). Multiple federal courts of appeals have held that, where citizenship rates are at issue in a vote-dilution case, citizen voting-age population is the proper metric for determining whether a racial group could constitute a majority in a single – member district. See, e.g., Reyes v. City of Farmers Branch, 586 F.3d 1019, 1023-24 (5th Cir. 2009); Barnett v. City of Chicago, 141 F.3d 699, 704 (7th Cir. 1998); Negron v. City of Miami Beach, 113 F.3d 1563, 1567-69 (11th Cir. 1997); Romero v. Pomona, 883 F.2d 1418, 1426 (9th Cir. 1989), overruled in part on other grounds by Townsend v. Holman Consulting Corp., 914 F.2d 1136 (9th Cir. 1990). A more accurate decennial census, which should collect this data, would prevent the inevitable dilution of votes and further exposure of the states to endless litigation. The current use of ACS data puts States into a no-win situation where they cannot apportion representative districts in a manner that is consistent with the Constitution, the Voting Rights Act, or their own state redistricting laws. It is clear that the intent of Section 2’s prohibition “is to facilitate participation. . .in our political process” by preventing unlawful vote dilution on account of race. Campos v. City of Houston, 113 F.3d 544, 548 (5th Cir. 1997). Courts have reasoned that “[t]he right to vote is one of the badges of citizenship” and that “[t]he dignity and very concept of citizenship are diluted if noncitizens are allowed to vote.” Barnett, 141 F.3d at 704. Thus, it would be the wrong result for a legislature or a court to draw a single-member

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 21051

district in which a numerical racial minority group in a jurisdiction was a majority of the total voting-age population in that district but “continued to be defeated at the polls” because it was not a majority of the citizen voting-age population. Campos, 113 F.3d at 548. As these cases show, for the U.S. Department of Justice to avert racial discrimination in voting and ensure adherence to the spirit of Section 2, it is vital that the “long form” citizenship question in the decennial Census provide the necessary citizen voting-age population data. When the right of all citizens to cast a properly weighted vote is not protected, there is a resulting dilution of the voting power of citizens residing in districts that are home to a smaller number of nonvoting residents. Moreover, it incentivizes sanctuary cities by granting an electoral advantage at the expense of non-sanctuary cities. Voting is one of the most precious rights of citizenship. And yet, it is clear that representative districts with larger populations of illegal and non-voting legal immigrants have gained representation over those without. Accordingly, I am imploring the Census Bureau to reinstate the citizenship question in the decennial 2020 Census to assist Louisiana and all other states in making a good faith effort to equalize districts in a method that ensures, as far as practicable, equality in the weight of votes. Sincerely, /s/ Jeff Landry Louisiana Attorney General

78228 • ACLU • APPENDIX 01 (DOCUMENTS 1 THRU 6) AL 3/21/19 21152

February 12, 2018

The Honorable Wilbur Ross Secretary of Commerce U.S. Department of Commerce 1401 Constitution Avenue NW Washington, DC 20230

Dear Secretary Ross, We, the undersigned Attorneys General of New York, Massachusetts, California, Connecticut, Delaware, the District of Columbia, Hawaii, Illinois, Iowa, Maine, Maryland, Mississippi, New Jersey, New Mexico, Oregon, Pennsylvania, Rhode Island, Vermont, and Washington, as well as the Governor of Colorado, write to oppose the recent request by the Department of Justice to add a question on citizenship to the questionnaire for the 2020 decennial Census.1 Adding a citizenship

1 See Letter from Arthur E. Gary, General Counsel, Justice Management Division, U.S. Dep’t of Justice, to Ron Jarmin, Performing the Non-Exclusive Functions and Duties of the Director, U.S. Bureau of the Census, U.S. Dep’t of Commerce (Dec. 12, 2017), https://www.documentcloud.org/ documents/4340651-Text-of-Dec-2017-DOJ-letter-to-Census. html [hereinafter DOJ Letter]. The Justice Department’s request that the Bureau “reinstate” a citizenship question on the Census, see id. at 1, is misleading, as no citizenship question has been included on the decennial census since 1950. From 1970 to 2000, a citizenship question was included only on the “long form” questionnaire, which was distributed to a sample of about one in six households in lieu of the decennial census questionnaire. Following the 2000 Census, the Census Bureau discontinued the “long form” questionnaire and replaced it with the American Community Survey, which is now sent to about one in every 38 households each year.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 21253

question – especially at such a late date in the 2020 Census planning process – would significantly depress participation, causing a population undercount that would disproportionately harm states and cities with large immigrant communities. This undercount would frustrate the Census Bureau’s obligation under the Constitution to determine “the whole number of persons in each state,”2 threaten our states’ fair representation in Congress, dilute our states’ role in the Electoral College, and deprive our states of their fair share of hundreds of billions of dollars in federal funds that are allocated in part on decennial Census data. Indeed, as the Census Bureau has itself previously explained, “any effort to ascertain citizenship” in the decennial Census “will inevitably jeopardize the overall accuracy of the population count.”3 These tremendous harms are not justified by the Justice Department’s purported interest in strengthening enforcement of Section 2 of the Voting Rights Act. To the contrary, requesting citizenship data would undermine the purposes of the Voting Rights Act and weaken voting rights enforcement across the board. For these reasons, we have serious concerns that adding a citizenship question to the 2020 Census at this late date would violate the Census Bureau’s obligations under the Constitution, the Administrative Procedure Act, and other federal statutes.

2 U.S. Const. amend. XIV, § 2; see also id. art. I, § 2, cl. 3. 3 Fed’n for Am. Immigration Reform v. Klutznick, 486 F. Supp. 564, 568 (D.D.C. 1980).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 21354

Furthermore, the underfunding of the Census Bureau raises concerns that technology and implementation strategies will not be adequately developed before the start of the full 2020 Census. The lack of testing in rural areas is particularly disconcerting. We request your assurances that the Bureau will be able to cope with this funding crisis and provide a full and accurate enumeration of the population of each state. I. Adding a citizenship question at this late date would fatally undermine the accuracy of the 2020 Census, harming the states and our residents. The Justice Department’s request should be rejected because adding a citizenship question to the 2020 Census would reduce participation and response rates, threatening the Census Bureau’s ability to comply with its obligations under the Constitution and harming the states’ interests. 1. Questions about citizenship would deter participation in the 2020 Census, undermining the constitutional mandate to conduct an “actual Enumeration.” The Constitution provides that Representatives “shall be apportioned among the several States . . . according to their respective Numbers,”4 which requires “counting the whole number of persons in each State.”5 This count is to be determined by an “actual Enumeration” conducted every ten years.6 It is well-settled that

4 Id. art. I, § 2, cl. 3. 5 Id. amend. XIV, § 2. 6 Id. art. I, § 2, cl. 3; see also 13 U.S.C. § 4 (delegating to the Secretary of Commerce authority to conduct the decennial census).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 21455

this “actual Enumeration” includes all residents, both citizens and noncitizens.7 A citizenship question would hinder the Census Bureau’s ability to complete this “actual Enumeration” by chilling participation in the 2020 Census by noncitizens and naturalized citizens alike. The Census Bureau has long recognized the difficulty of counting immigrant and noncitizen communities. In preparing for the 2010 Census, the Bureau identified immigrants as one of several hard-to-count populations, and designed a significant public education campaign to increase participation from that group.8 Similarly, in the lead up to the current decennial Census, the Bureau organized a working group to recommend strategies to minimize undercounts of undocumented immigrants, as well as immigrant Latinos and Asians.9 Notwithstanding these efforts, the difficulty of counting such groups has only increased in the current climate. Recent pretests by the Census

7 Klutznick, 486 F. Supp. at 575-77. 8 U.S. Census Bureau, 2010 Census Integrated Communications Plan, 21, 75, 191, 223, 278 (Aug. 2008), https://www.census.gov/2010census/partners/pdf/2010_ICC_ Plan_Final_Edited.pdf; U.S. Gov’t Accountability Office, GAO-09-525T, Communications Campaign Has Potential to Boost Participation, 6 (Mar. 6, 2009), https://www.gao.gov/ assets/130/122012.pdf. 9 U.S. Census Bureau, Nat’l Advisory Comm. on Racial, Ethnic, and Other Populations, Final Report of the Administrative Records, Internet, and Hard to Count Population Working Group, 2, 8 (July 2016), https://www2. census.gov/cac/nac/reports/2016-07-admin_internet-wg-report. pdf.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 21556

Bureau have revealed that immigrant respondents increasingly expressed concerns about confidentiality and data sharing, especially when asked questions about citizenship.10 Citing fears related to the current discourse on immigration policy, respondents have also refused to respond to questions and have ended interactions with surveyors.11 The Census Bureau has recognized that these anxieties might present a barrier to participation in the 2020 Census, and may diminish overall data quality.12 Even before the Department of Justice made its request, Census Bureau officials reported that early test surveys showed “an unprecedented groundswell in confidentiality and data-sharing concerns among immigrants or those who live with immigrants” related to the 2020 count.13 The Bureau already

10 Memorandum from the U.S. Census Bureau, Ctr. for Survey Measurement, to Assoc. Directorate for Research and Methodology, 1, 5-7 (Sept. 20, 2017), https://www2.census. gov/cac/nac/meetings/2017-11/Memo-Regarding-Respondent- Confidentiality-Concerns.pdf. 11 Id. at 2. 12 U.S. Census Bureau, Nat’l Advisory Comm. on Racial, Ethnic, and Other Populations, Respondent Confidentiality Concerns and Possible Effects on Response Rates and Data Quality for 2020 Census, 2, 12-13, 15 (Nov. 2, 2017), https://www2.census.gov/cac/nac/meetings/2017- 11/Meyers-NAC-Confidentiality-Presentation.pdf. 13 Mica Rosenberg, U.S. Officials Worry Immigrant Fears Could Make Census Inaccurate, Reuters (Nov. 30, 2017, 3:10 PM), https://www.reuters.com/article/us-usa- immigration-census/u-s-officials-worry-immigrant-fears-could- make-census-inaccurate-idUSKBN1DU2U7. These concerns from some federal officials are shared by state-level experts with experience coordinating the administration of the decennial Census in their states. Massachusetts Secretary of

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 21657

acknowledges that questions about citizenship in any federal statistical survey are sensitive and must be treated with care14; adding a citizenship inquiry to the mandatory decennial Census would undoubtedly exacerbate these problems, leading to larger undercounts and less reliable data. Indeed, in a brief filed with the Supreme Court less than three years ago, four former Directors of the Census Bureau – appointed by Presidents of both political parties – explained based on their experience that “a one-by-one citizenship inquiry would invariably lead to a lower response rate to the Census in general,” and would “seriously frustrate the Census Bureau’s ability to conduct the only count the Constitution expressly requires: determining the whole number of persons in each state in order to apportion House seats among the states.”15 The former Directors explained that “[r]ecent experience demonstrates lowered participation in the Census and increased suspicion of government collection of information in general.

the Commonwealth, William Galvin, for example, recently testified before a state legislative committee that a citizenship inquiry would be a clear deterrent to participating in the 2020 Census. See Christina Prignano, Mass. secretary of state warns Trump could “sabotage” 2020 Census, Boston Globe (Feb. 6, 2018), https://www.bostonglobe.com/metro/ 2018/02/06/mass-secretary-state-warns-trump-could-sabotage- census/HH2b73v0o2dkddzrjYDyUK/story.html. 14 U.S. Census Bureau, Data Stewardship Exec. Policy Comm., DS-16: Policy on Respondent Identification and Sensitive Topics in Dependent Interviewing, 1-2 (Dec. 9, 2014), https://www2.census.gov/foia/ds_policies/ds016.pdf. 15 Brief of Former Directors of the U.S. Census Bureau as Amici Curiae Supporting Appellees at 25, Evenwel v. Abbott, 136 S. Ct. 1120 (2016) (No. 14-940).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 21758

Particular anxiety exists among non-citizens. There would be little incentive for non-citizens to offer to the government their actual status; the result [of inquiring about citizenship status] would be a reduced rate of response overall and an increase in inaccurate responses.”16 The Census Bureau in fact declined to add a citizenship question to the 2010 Census questionnaire,17 and has repeatedly warned against adding such a question to the decennial Census because of the risk of lower response rates and reduced accuracy.18 As the Census Bureau has explained, questions about “citizenship are particularly sensitive” for individuals who “perceive[] any possibility of the information being used against them,” and thus “any effort to ascertain citizenship will inevitably jeopardize the overall accuracy of the population count” required by the Constitution.19 2. This threat to the accuracy of the 2020 Census is magnified by the extreme lateness of the Justice

16 Id. at 5. 17 See U.S. Census Bureau, 2010 Census Memorandum Planning Series No. 239, 2010 Census Content and Forms Design Program Assessment Report, 14 (Sept. 25, 2012). 18 See Census Equity Act: Hearings Before the Subcomm. on Census & Population of the Comm. on Post Off. & Civ. Serv. 43-45 (1989) (statement of C. Louis Kincannon, Deputy Director, U.S. Bureau of the Census); Exclude Undocumented Residents from Census Counts Used for Apportionment: Hearing Before the Subcomm. on Post Office & Civil Serv., 100th Cong. 50-51 (1988) (testimony of John Keane, Director, U.S. Bureau of the Census) [hereinafter Census Counts]. 19 Klutznick, 486 F. Supp. at 568.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 21859

Department’s proposal. Even assuming it were possible to devise a citizenship inquiry that would not risk an unconstitutional undercount, it is far too late in the planning process for the Census Bureau to test and validate any such approach. The Bureau must meet a statutory deadline of March 31, 2018 – less than two months away – to submit its final questionnaire for the 2020 Census to Congress.20 Two months is insufficient time to design and test a question as sensitive as this one consistent with the guidelines that apply to federal statistical agencies. By statute, the Office of Management and Budget (OMB) has responsibility for coordinating the federal statistical system, including to ensure “the integrity, objectivity, impartiality, utility, and confidentiality of information collected for statistical purposes.”21 OMB is also required to establish government-wide guidelines and policies regarding statistical collection methods.22 Consistent with these statutory obligations, OMB has published a number of Statistical Policy Directives that govern

20 13 U.S.C. § 141(f)(2) (providing, with respect to each decennial census, “the Secretary [of Commerce] shall submit to the committees of Congress having legislative jurisdiction over the census . . . not later than 2 years before the appropriate census date, a report containing the Secretary’s determination of the questions proposed to be included in such a census”); 13 U.S.C. § 141(a) (establishing April 1, 2020 as the decennial census date). 21 44 U.S.C. § 3504(e)(1); see also 44 U.S.C. § 3501(9); 44 U.S.C. §§ 3504(a)(1)(B)(iii), (e); Office of Mgmt. & Budget, Statistical Programs of the United States Government, Fiscal Year 2017, 3-4, 11 (2017). 22 44 U.S.C. § 3504(e)(3).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 21960

the data collection efforts of federal statistical agencies, including the Census Bureau.23 These guidelines require, among other obligations, that agencies “ensure that all components of a survey function as intended . . . by conducting a pretest of the survey components or by having successfully fielded the survey components on a previous occasion.”24 OMB specifically recommends pretesting new components of a survey prior to a field test, and incorporating results into the final design. In addition, the Census Bureau has further clarified the statistical standards it must utilize to address the agency’s unique methodological and operational challenges.25 These standards require that all data collection instruments be tested “in a manner that balances data quality and respondent burden,” and specifically require pretesting to ensure questions are not “unduly sensitive” and “do not cause undue burden.”26

23 Office of Mgmt. & Budget, Statistical Policy Directive No. 1: Fundamental Responsibilities of Federal Statistical Agencies and Recognized Statistical Units, 79 Fed. Reg. 71,610 (Dec. 2, 2014); Office of Mgmt. & Budget, Statistical Policy Directive No. 2: Standards & Guidelines for Statistical Surveys (Sept. 2006); Office of Mgmt. & Budget, Statistical Policy Directive No. 4: Release and Dissemination of Statistical Products Produced by Federal Statistical Agencies, 73 Fed. Reg. 12,621 (Mar. 7, 2008). 24 Office of Mgmt. & Budget, Statistical Policy Directive No. 2, § 1.4 at 9 (2006). 25 See U.S. Census Bureau, Statistical Quality Standards, ii (Jul. 2013), https://www.census.gov/content/ dam/Census/about/about-the-bureau/policies_and_notices/quality/ statistical-quality-standards/Quality_Standards.pdf. 26 Id. at 7-8 reqs. A2-3 & A2-3.3.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22061

These requirements cannot reliably be met in the limited time available before the Census Bureau’s March 31 deadline. The Census Bureau already developed and approved its National Content Test in 2015, which it characterized as its “primary mid- decade opportunity to compare different versions of questions prior to making final decisions for the 2020 Census.”27 And the 2018 End-to-End Census Test – which the Census Bureau describes as the “culmina- tion” of its years-long process of testing and validating all aspects of the decennial Census design – is already underway, having begun in August 2017.28 In short, there is insufficient time for the Census Bureau to conduct the extensive development and testing that would be required to comply with OMB guidelines for adding new questions to the 2020 Census while assuring its validity and accuracy. And as the Census Bureau has explained, conducting the Census with “untested and unproven procedures” would further undermine the Bureau’s ability to conduct “a timely, accurate” enumeration.29 These concerns are heightened even further by the Census Bureau’s already-precarious fiscal position as it prepares for the 2020 Census. The Bureau is dramatically underfunded, and the addition of a citizenship question would add

27 U.S. Census Bureau, Information Collection Request: 2015 National Content Test, 80 Fed. Reg. 29,609, 29,610 (May 22, 2015). 28 U.S. Census Bureau, Frequently Asked Questions for the 2018 End-to-End Census Test (Dec. 20, 2017), https:// www.census.gov/programs-surveys/decennial-census/2018-cen sus-test/faqs.html. 29 Census Counts, at 49-50.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22162

significantly to the overall price of completing the Census. The Bureau’s appropriated budget for Fiscal Year 2017 was roughly ten percent below its request, and was finalized seven months late.30 And the administration’s initial budget request for Fiscal Year 2018 proposed only a two percent increase for the Census Bureau over the previous year – well short of the resources needed for the Bureau to prepare adequately for the decennial Census.31 Further exacerbating these budget constraints, the reduced response rates that a citizenship question would cause will result in vastly increased costs overall. Reduced response rates trigger an expensive in-person follow-up process, which could result in an estimated increase of hundreds of millions of dollars to the price tag for the 2020 Census. Because of inadequate financial resources, unreliable cost estimates, information technology challenges, and other concerns, GAO has already placed the 2020 Census on its “High Risk List” of government programs at greatest risk of fraud, waste, abuse, and mismanagement.32 Adding the

30 Robert Shapiro, The 2020 Census May Be Wildly Inaccurate – And It Matters More Than You Think, BROOKINGS (Aug. 31, 2017), https://www.brookings.edu/blog/ fixgov/2017/08/31/the-2020-census-may-be-wildly-inaccurate- and-it-matters-more-than-you-think/. 31 See id. (noting that the Census Bureau’s funding increased 60 percent between 2007 and 2008 in advance of the 2010 Census). 32 U.S. Gov’t Accountability Office, GAO-17-317, High- Risk Series: Progress on Many High-Risk Areas, While Substantial Efforts Needed on Others, 220-31 (Feb. 2017), https://www.gao.gov/assets/690/682765.pdf.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22263

challenge of testing and validating a question on citizenship to the tremendous operational and planning challenges that the Census Bureau already faces would increase the risk of error and heighten the chance of an undercount in our states. 3. The states would be irreparably harmed by an inaccurate 2020 Census. By deterring participation in the Census, the proposed citizenship question would harm everyone, citizens and noncitizens alike. First, an inaccurate 2020 Census could result in widespread malapportionment of the states’ representation in Congress. As noted, the Constitution requires that Representatives “shall be apportioned among the several States . . . according to their respective Numbers.”33 As provided by the Census Act, the Secretary of Commerce is required to use the decennial Census results to tabulate the total population by state and report those results to the President,34 who must then “transmit to the Congress a statement showing the whole number of persons in each State . . . and the number of Representatives to which each State would be entitled.”35 An undercount that fails accurately to report the “whole number of persons” in each state would result in an incorrect calculation of the number of Representatives to which each state is entitled, in violation of the Census Clause of the

33 U.S. Const. art. I, § 2, cl. 3. 34 13 U.S.C. § 141(a). 35 2 U.S.C. § 2a(a).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22364

Constitution.36 Inaccurate data would also jeopardize the ability of the states – and all of our local jurisdictions – to comply with the Fourteenth Amendment’s one-person one-vote requirement when drawing district lines for everything from the state legislature to local city councils.37 Moreover, there would be no possibility of correcting this harm for at least a decade, when the next decennial Census takes place – and no way to undo the harm the states would suffer from a ten-year deprivation of their constitutional allotment of Representatives. In addition, a Census undercount could affect state representation in the Electoral College. The Constitution assigns each state a number of electors equal to “the whole number of Senators and Representatives to which the State may be entitled in the Congress.”38 An undercount that affected the apportionment of Representatives would also misrepresent the number of electors each state should receive, thereby miscalculating

36 See, e.g., Utah v. Evans, 536 U.S. 452, 459 (2002) (challenge by the State of Utah and its Congressional delegation to a Census Bureau methodology that resulted in Utah receiving one less Representative in Congress); Franklin v. Massachusetts, 505 U.S. 788, 790-91 (1992) (challenge by the Commonwealth of Massachusetts to the Census Bureau’s change in the method of counting overseas federal employees, which caused Massachusetts to receive one less seat in the House of Representatives). 37 See Reynolds v. Sims, 377 U.S. 533, 568 (1964); Baker v. Carr, 369 U.S. 186, 208-09, 237 (1962). 38 U.S. Const. art. II, § 2, cl. 2; see also id. amend. XII, amend. XXIII (allocating electors to the District of Columbia).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22465

each state’s proper role in selecting the President and Vice President. This extraordinary harm to the fabric of our federal system would come with equally significant financial harm. Data derived from the decennial Census guide the geographic distribution of hundreds of billions of dollars in federal grant funds to states and local areas. According to one estimate, there are about 300 Census-guided federal grant programs, with total appropriations in Fiscal Year 2015 of approximately $700 billion.39 These programs include Medicaid, the Supplemental Nutritional Assistance Program (SNAP), Title I grants to local educational agencies under the Elementary and Secondary Education Act, formula grants for highway planning and construction, Section 8 housing choice vouchers, the Low-Income Home Energy Assistance Program, and more.40 In other words, a Census undercount would jeopardize critical federal funding the states need to provide health insurance, public education funding, food assistance, housing opportunities, energy assistance, and other services and support for millions of residents, regardless of citizenship status. Such widespread underfunding harms everyone, starting with the most vulnerable, including low- income communities and children.

39 See Andrew Reamer, Counting for Dollars 2020: The Role of the Decennial Census in the Geographic Distribution of Federal Funds, G.W. Inst. Pub. Pol’y (Aug. 22, 2017), https://gwipp.gwu.edu/counting-dollars-role-decennial-census- geographic-distribution-federal-funds. 40 See id.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22566

The Census Bureau has both constitutional and statutory obligations to conduct an “actual enumeration.” Including a question on the 2020 Census that would manipulate the count by scaring people away from being counted – causing grave harm to the states and our residents – is inconsistent with those obligations.41 II. Adding a citizenship question to the 2020 Census would hamper the goals of the Voting Rights Act. The Justice Department’s request for citizenship data asserts that this information is necessary to ensure compliance with Section 2 of the Voting Rights Act. In fact, voting rights compliance will be undermined – not enhanced – by the addition of a citizenship question to the 2020 Census. Because the Justice Department’s request is unsupported by its stated reason, adding a citizenship question would be arbitrary and capricious under the Administrative Procedure Act.42 1. Collecting citizenship data would undermine the goal of fair and effective representation for all communities, which the Voting Rights Act was enacted to protect. The purpose of the Voting

41 Cf. Dep’t of Commerce v. U.S. House of Representatives, 525 U.S. 316, 348 (1999) (Scalia, J., concurring) (noting that the purpose of a “genuine enumeration” is to accomplish “the most accurate way of determining population with minimal possibility of partisan manipulation”). 42 See Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (noting that an agency acts arbitrarily and capriciously when it “entirely fail[s] to consider an important aspect of the problem” or “offer[s] an explanation for its decision that runs counter to the evidence before the agency”).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22667

Rights Act is to accomplish “nondiscriminatory treatment by government – both in the imposition of voting qualifications and the provision or administration of governmental services, such as public schools, public housing and law enforcement.”43 Any method of enumeration that predictably undercounts some communities – as the Justice Department’s proposal would do – will mean that those communities are not fairly represented when legislative seats are apportioned and district lines are drawn. The Supreme Court has long made clear that legislators represent all constituents in the districts they serve, regardless of whether any particular individual is a citizen: “[T]he fundamental principle of representative government in this country” is “one of equal representation for equal numbers of people.”44 The Justice Depart- ment’s request should be rejected because it would undermine this fundamental principle. 2. Citizenship data from the decennial Census is unnecessary to enforce the vote-dilution prohibition in Section 2 of the Voting Rights Act. The Justice Department’s request should also be rejected because it is unsupported. The Justice

43 Katzenbach v. Morgan, 384 U.S. 641, 652 (1966). 44 Reynolds, 377 U.S. at 560-61; see also Evenwel v. Abbott, 136 S. Ct. 1120, 1131-32 (2016); Davis v. Bandemer, 478 U.S. 109, 132 (1986) (plurality opinion); Daly v. Hunt, 93 F.3d 1212, 1226 (4th Cir. 1996) (explaining that “people can affect what their representatives do in another way” besides voting: “through their right to petition their representatives to voice their concerns and interests on particular issues. This right is available to everyone, even those who are ineligible to vote.”).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22768

Department contends that it needs a “reliable calculation of citizen voting-age population” (or “CVAP”) in order to enforce the vote-dilution prohibition of Section 2.45 But the Supreme Court has never held that citizen voting-age population is the proper measure for examining whether a minority group can constitute a majority in a single- member district (the first element of proving a vote- dilution claim).46 The Justice Department notes that in LULAC v. Perry, the Supreme Court “analyz[ed] a vote-dilution claim by reference to citizen voting- age population,”47 but fails to note that in a subsequent Section 2 case – Bartlett v. Strickland – the Court assessed the vote-dilution inquiry in terms of “voting-age population.”48 The question of the appropriate population measure in Section 2 vote-dilution cases is, at best, unsettled.49

45 DOJ Letter at 1. 46 Thornburg v. Gingles, 478 U.S. 30, 50-51 (1986) 47 DOJ Letter at 1 (citing LULAC v. Perry, 548 U.S. 399, 423-442 (2006)). 48 Bartlett v. Strickland, 556 U.S. 1, 12 (2009) (“This case turns on whether the first Gingles requirement can be satisfied when the minority group makes up less than 50 percent of the voting-age population in the potential election district.”); see also id. at 18 (“Unlike any of the standards proposed to allow crossover-district claims, the majority- minority rule relies on an objective, numerical test: Do minorities make up more than 50 percent of the voting-age population in the relevant geographic area? That rule provides straightforward guidance to courts and to those officials charged with drawing district lines to comply with § 2.”). 49 See, e.g., Sanchez v. State of Colo., 97 F.3d 1303, 1311 (10th Cir. 1996) (“Because Gingles advances a functional evaluation of whether the minority population is large enough to form a district in the first instance, the Circuits have been flexible in assessing the showing made for this precondition.”).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22869

In addition, even if citizen voting-age population were required in all cases, adding a citizenship question to the Census would not give the Justice Department the “reliable calculation” of citizenship information it claims to need. The Census is of course only administered every ten years,50 so any CVAP figures from the decennial Census would quickly become outdated and less reliable over the course of the subsequent decade as a result of population shifts. And a citizenship question would not provide information sufficient to ascertain the precise number of eligible voters in a district because district residents might be ineligible to vote for other reasons, such as prior felony convictions. In any event, the Census Bureau’s American Community Survey already collects citizenship data, and these estimates are available for the federal government to use as needed. Indeed, Congress could not possibly have intended for effective Section 2 enforcement to depend on the availability of person-by-person citizenship data, because such data has never been available at any point since Section 2 has existed: not in 1965 when the Voting Rights Act was first enacted; not in 1982 when the Act was amended to clarify the vote-dilution standard; not in 1986 when the Supreme Court articulated the vote-dilution test in Thornburg v. Gingles. Because the Justice Department’s request seeks data that has never before been required in Section 2 litigation – and that cannot reliably be collected in

50 U.S. Const. art. I, § 2, cl. 3; 13 U.S.C. § 141(a).

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 22970

any event – it cannot credibly serve as the basis for major changes to the 2020 Census design that will undercut the accuracy of the constitutionally mandated enumeration. III. The addition of a question regarding citizenship to the 2020 Census is inconsistent with the Census Bureau’s Information Quality Guidelines. The Information Quality Act (“IQA”) requires agencies to ensure that the information they disseminate to the public is accurate, reliable, and objective.51 Consistent with this directive, the IQA requires OMB and other federal agencies to issue guidelines “ensuring and maximizing the quality, objectivity, utility, and integrity of information, including statistical information, disseminated by the agency.”52 Recognizing the critical importance of the information it disseminates, the Census Bureau has adopted particularly stringent agency-specific IQA guidelines. These guidelines provide detailed requirements that the Census Bureau must meet to ensure the “utility,” “objectivity,” “integrity,” and “transparency” of information from the decennial Census.53

51 Consolidated Appropriations Act, 2001, Pub. L. No. 106-554, § 515, 114 Stat. 2763 (Dec. 21, 2000). 52 Id.; see also Guidelines for Ensuring and Maximizing the Quality, Objectivity, Utility and Integrity of Information Disseminated by Federal Agencies, 67 Fed. Reg. 8457 (Feb. 22, 2002). 53 Information Quality Guidelines, U.S. Census Bureau (May 12, 2015), https://www.census.gov/about/policies/ quality/guidelines.html.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23071

The Census Bureau’s IQA guidelines disfavor questions that diminish response rates. The Bureau’s guideline for ensuring “objectivity,” requires collection and dissemination of information that is “accurate, reliable and unbiased.”54 To achieve this end, the guideline requires the Census Bureau to utilize collection methods that “minimiz[e] respondent burden.”55 This concern recognizes that respondents may choose not to respond when confronted by a question that is unduly sensitive or burdensome.56 Burdensome questions may diminish the accuracy and reliability of data collected in surveys by driving down response rates. Indeed, the Census Bureau has acknowledged this very concern by adopting statistical standards that test for and revise these types of questions.57 The addition of a question regarding citizenship will diminish overall response rates. As noted above, many immigrant and citizen groups are likely to be highly sensitive to the citizenship inquiry. Adding this question to the 2020 Census questionnaire would impose a high burden on

54 Information Quality: Objectivity, U.S. Census Bureau (Apr. 17, 2015), https://www.census.gov/about/policies/ quality/guidelines/objectivity.html. 55 Id; Similarly, OMB’s statistical standards require the Census Bureau to design its data collection instruments and methods “in a manner that achieves the best balance between maximizing data quality . . . while minimizing respondent burden and cost.” Office of Mgmt. & Budget, Statistical Policy Directive No. 2, § 2.3 at 11. 56 U.S. Census Bureau, Statistical Quality Standards, at A2-3.3. 57 Id.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23172

these groups, dissuade many from responding, and impair the survey’s ultimate accuracy and reliability. As a result, by adding a citizenship inquiry to the questionnaire, the Census Bureau would hinder compliance with its own objectivity standard. Moreover, the Census Bureau has not taken any steps to test the citizenship inquiry and its impact on potential respondents. The objectivity standard applies not only to the utilization of a particular data collection method, but also to the development of that method.58 As noted above, both OMB and the Census Bureau have adopted statistical standards that require pre-testing in the development of data collection methods and survey questions.59 To date, the Census Bureau has not engaged in any pretesting of the citizenship question. As a result, adoption of the citizenship question would conflict with the agency’s IQA guidelines, and the Census Bureau should reject requests to include that question on the 2020 Census questionnaire. IV. Conclusion. Fair, proportionate electoral representation in our democracy depends on valid Census data. The proposal to add a citizenship question to the 2020 Census questionnaire would defeat that goal, violate the Constitution, and undermine the purposes of the Voting Rights Act that the Justice Department claims it wants to protect. Because inclusion of a citizenship

58 U.S. Census Bureau, Information Quality: Objectivity. 59 Office of Mgmt. & Budget, Statistical Policy Directive No. 2, § 1.4 at 9; U.S. Census Bureau, Statistical Quality Standards, ii.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23273

question would threaten the Census Bureau’s ability to conduct its constitutionally-mandated role, and would be arbitrary and capricious under the Administrative Procedure Act – causing significant, direct harm to our states and residents – we urge you to reject the Justice Department’s request. Sincerely, /s/ Eric T. Schneiderman ERIC T. SCHNEIDERMAN Attorney General of the State of New York /s/ Maura Healey MAURA HEALEY Attorney General for the Commonwealth of Massachusetts /s/ Xavier Becerra XAVIER BECERRA Attorney General of the State of California /s/ John W. Hickenlooper JOHN W. HICKENLOOPER Governor of the State of Colorado /s/ George Jepsen GEORGE JEPSEN Attorney General of the State of Connecticut /s/ Matthew Denn MATTHEW DENN Attorney General of the State of Delaware /s/ Karl A. Racine KARL A. RACINE Attorney General for the District of Columbia

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23374

/s/ Russell Suzuki RUSSELL SUZUKI Acting Attorney General of the State of Hawaii /s/ Lisa Madigan LISA MADIGAN Attorney General of the State of Illinois /s/ Thomas J. Miller THOMAS J. MILLER Attorney General of the State of Iowa /s/ Janet T. Mills JANET T. MILLS Attorney General of the State of Maine /s/ Brian Frosh BRIAN FROSH Attorney General of the State of Maryland /s/ Jim Hood JIM HOOD Attorney General of the State of Mississippi /s/ Gurbir Grewal GURBIR GREWAL Attorney General of the State of New Jersey /s/ Hector H. Balderas HECTOR H. BALDERAS Attorney General of the State of New Mexico /s/ Ellen F. Rosenblum ELLEN F. ROSENBLUM Attorney General of the State of Oregon /s/ Josh Shapiro JOSH SHAPIRO Attorney General of the Commonwealth of Pennsylvania

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23475

/s/ Peter Kilmartin PETER KILMARTIN Attorney General of the State of Rhode Island /s/ Thomas J. Donovan, Jr. THOMAS J. DONOVAN, JR. Attorney General of the State of Vermont /s/ Bob Ferguson BOB FERGUSON Attorney General of the State of Washington cc: The Honorable Mick Mulvaney Director, Office of Management and Budget Arthur E. Gary General Counsel, Justice Management Division U.S. Department of Justice Dr. Ron Jarmin Performing the Non-Exclusive Functions and Duties of the Director U.S. Bureau of the Census

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23576

PRIVILEGED AND CONFIDENTIAL PRE-DECISIONAL DRAFT As part of his decision-making process, Secretary Ross spoke to a number of different stakeholders about the Department of Justice’s request to reinstate the citizenship question on the 2020 Decennial. These notes attempt to memorialize those conversations. These are not verbatim transcripts and each summary reflects the recollections of attendees from the Department of Commerce. Every effort has been made to ensure these notes are an accurate reflection of Secretary Ross’s conversations with stakeholders. Hermann Habermann, former Deputy Director and COO of the Census Bureau (2002-2006) On March 23, 2018, Secretary Ross and his staff spoke with Hermann Habermann, former Deputy Director and COO of the Census Bureau, former Director of the U.N. Statistical Division, and former Chief Statistician at OMB. Mr. Habermann stated that he was not aware of a controlled study that could quantify the effect on participation rates of asking a citizenship question. Mr. Habermann stated that he believed that asking a citizenship question on the Decennial Census would diminish response rates and degrade the quality of responses, but there is no data to support these beliefs or to quantify the expected response diminution rate. Mr. Habermann stated that he believed the “burden of proof” for getting a question added to the Decennial Census is on the person who proposes it. Specifically, the proposing

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23677

party should be required to demonstrate how the proposed question would not degrade the census. Mr. Habermann stated that the census is fragile, and that it is particularly fragile now because our country is divided and people are influenced by social media, which can be a powerfully disrupting force. Mr. Habermann continued that social media makes it much easier to galvanize mistrust about the census by questioning its very purpose. Mr. Habermann stated that lower response rates cause the costs of the census to go up and the quality of the data to go down. Mr. Habermann shared an example from his time at Census Bureau. In 2004, DHS asked the Census Bureau to provide data on the number of Arab Americans by zip code in certain areas of the country. Mr. Habermann noted that this information was already available to the public but DHS could not figure out how to access it. When the Census Bureau provided DHS with the information it requested, there was a political firestorm and the Census Bureau was accused of providing DHS with sensitive information. (Mr. Habermann made clear that the Census Bureau does not give out personally identifiable information and did not do so here, but the result was the same.) Mr. Habermann noted that despite the outcry, the response rate to subsequent census surveys did not change in the communities most impacted by the dissemination of the supposedly sensitive information. Mr. Habermann confirmed that he ascertained this personally, but also cautioned that we are living in a different time now and the political climate is different.

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23778

Mr. Habermann stated that he believed that reinstating a citizenship question would cause divisiveness and that the party requesting the addition should have the burden of proof to establish the overriding policy reason for the addition. Mr. Habermann further stated that if the Secretary wants to add the question, the reason must be clear – there must be no public mistrust of the underlying reason, which is not the case here. Mr. Habermann noted that this proposed citizenship question would be particularly fraught because there has not been a clear explanation given as to why this data is necessary. Therefore, it is easy to misconstrue the motives behind the question. Finally, Mr. Habermann noted that if a proposed question would not decrease cost, serve an important policy objective, or increase data quality, there is no reason to put it on the questionnaire. • Lower response rate • Degrade quality of responses • Burden of proof on proposing party • Country divided • Higher costs

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23879

PRIVILEGED AND CONFIDENTIAL PRE-DECISIONAL DRAFT As part of his decision-making process, Secretary Ross spoke to a number of different stakeholders about the Department of Justice’s request to reinstate the citizenship question on the 2020 Decennial. These notes attempt to memorialize those conversations. These are not verbatim transcripts and each summary reflects the recollections of attendees from the Department of Commerce. Every effort has been made to ensure these notes are an accurate reflection of Secretary Ross’s conversations with stakeholders. Christine Pierce, SVP of Data Science, Nielsen On March 23, 2018, Secretary Ross and his staff spoke with Christine Pierce, Senior Vice President of Data Science for Nielsen. Ms. Pierce shared that Nielsen uses census data in a lot of important ways, specifically how they recruit and project samples. Ms. Pierce stated that Nielsen needed the census to be accurate and needed the census to be efficient and that the best census is one that produces the highest quality data at the lowest cost. Ms. Pierce stated that her biggest concerns was that the reinstatement of a citizenship question could lead to a lower response rate, and that the mailback rate (or initial response rate) is very important. Costs are lower when people respond the first time. Failure to respond increases costs because Census Bureau needs to deploy enumerators. Ms. Pierce stated that including a question on citizenship could make people less likely to respond,

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 23980

but that there is no data to predict how much lower the response rate might be. In response to a question, Ms. Pierce stated that the longer a survey is, the less likely people are to respond. She further stated that the more sensitive the question, the more likely people are to be turned off by the question and decline to respond. Ms. Pierce explained that examples of sensitive questions included questions or religion and sexuality. Ms. Pierce stated that Nielsen sometimes chooses to ask sensitive questions even if they believe it will depress response rates. Ms. Pierce stated that Nielsen conducts a cost-benefit analysis to determine whether it is worth asking the question, even if it means having to do more extensive nonresponse follow-up. Ms. Pierce stated that sensitive questions often appeared on longer surveys and that longer surveys generally had lower response rates than shorter ones. Ms. Pierce stated that she was not aware of a short census survey that contained a sensitive question, but that Nielsen has tested some of the ACS questions perceived to be “sensitive” (birthplace and date of arrival in the US) on shorter surveys. Ms. Pierce noted that she and others at Nielsen were concerned about response rates declining due to the presence of the sensitive questions on the short questionnaire, but that Nielsen did not observe lower response rates to the survey. Ms. Pierce noted the importance of testing questions. She also noted that in the only specific situation she was aware of that sensitive questions were tested on a short questionnaire, there was no impact on response rates. Finally, in response to a question, Ms. Pierce stated that Nielsen incentivize

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 24081

participation with low dollar cash reward in the $1-$15 range. Ms. Pierce believed that for the survey referenced above, any incentive would have been at the lower end of the range. • Lower response rate/higher NRFU • Higher costs • Testing

78228 • ACLU • APPENDIX 02 (DOCS 7 THRU 9) AL 3/21/19 24182

PX-045

From: Wilbur Ross [REDACTED] Sent: 9/1/2017 3:12:16 AM To: Comstock, Earl (Federal) [REDACTED] CC: Teramoto, Wendy (Federal) [REDACTED] Subject: Re: [REDACTED] I have received no update, nor has there been an update [REDACTED], nor the issue of the census question, nor whether KDB thinks we have our arms around the census cost data nor another candidate. To run census, [REDACTED] Sent from my iPad > on Aug 31, 2017, at 6:29 AM, Comstock, Earl (Federal) [REDACTED] wrote: > [REDACTED] > On 8/30/17, 10:37 PM, “Wilbur Ross [REDACTED] wrote: [REDACTED] Earl > > Sent from my iPad > >> on Aug 30, 2017, at 5:47 PM, Comstock, Earl (Federal) [REDACTED] wrote: >> [REDACTED]

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 24283

>> From: “Comstock, Earl (Federal)” [REDACTED] >> Date: Wednesday, August 30, 2017 at 5:44 PM >> To: “Ross, Wilbur (Federal)” [REDACTED] >> CC: Wendy Teramoto [REDACTED] >> Subject: [REDACTED] >> >> Mr. Secretary - [REDACTED] >> Thank you. >> >> Earl >> >> [cid:[email protected]] >> [cid:[email protected]] >> [FU Scansnap Manager #iXSOO] >> >> >> >> >> >>

[STAMP] EXHIBIT 23 WIT: COMSTOCK DATE: 8/30/18 KLJ

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 24384

PX-048

To: Teramoto, Wendy (Federal) [REDACTED]@doc.gov] From: Comstock, Earl (Federal) Sent: Sat 9/16/2017 11:33:38 AM Importance: Normal Subject: Calls with DoJ Received: Sat 9/16/2017 11:33:38 AM Morning Wendy – Here is the memo I gave SWLR regarding my discussions with DoJ. Earl *** September 8, 2017 To: Secretary Wilbur Ross Fr: Earl Comstock Re: Census Discussions with DoJ In early May Eric Branstad put me in touch with Mary Blanche Hankey as the White House liaison in the Department of Justice. Mary Blanche worked for AG Sessions in his Senate office, and came with him to the Department of Justice. We met in person to discuss the citizenship question. She said [REDACTED]. A few days later she directed me to James McHenry in the Department of Justice. I spoke several times with James McHenry by phone, and after considering the matter further James said [REDACTED]. James directed me to Gene Hamilton at the Department of Homeland Security.

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 24485

Gene and I had several phone calls to discuss the matter, and then Gene relayed that after discussion DHS really felt that it was best handled by the Department of Justice. At that point the conversation ceased and I asked James Uthmeier, who had by then joined the Department of Commerce Office of General Counsel, to [REDACTED]. [STAMP] EXHIBIT Teramoto 7 8/24/18 TD

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 24586

PX-055

To: Wilbur Ross [REDACTED] Cc: Branstad, Eric (Federal) [[email protected]] From: Comstock, Earl (Federal) Sent: Fri 3/10/2017 8:31:29 PM Importance: Normal Subject: Your Question on the Census Received: Fri 3/10/2017 8:31:30 PM I was not able to catch anyone at their desk when I called the numbers I have for the Census Bureau from their briefing. However, the Census Bureau web page on apportionment is explicit and can be found at https:// www.census.gov/population/apportionment/ about/faq.html#Q16 It says: Are undocumented residents (aliens) in the 50 states included in the apportionment population counts? Yes, all people (citizens and noncitizens) with a usual residence in the 50 states are to be included in the census and thus in the apportionment counts. Further, this WSJ blog post from 2010 confirms that neither the 2000 nor the 2010 Census asked about citizenship. http://blogs.wsj.com/numbers/the-pitfalls-of-count ing-illegal-immigrants-937/ [STAMP] EXHIBIT 2 WIT: Comstock DATE: 8/30/18 KLJ

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 24687

THE NUMBERS The Pitfalls of Counting Illegal Immigrants [IMAGE] By CARL BIALIK May 7, 2010 7:05 pm ET The debate over Arizona’s immigration law has included several estimates of the state’s illegal- immigrant population, at “almost half a million,” “half a million” or “more than half a million.” Arguing against the law, Homeland Security chief Janet Napolitano — who is the former governor of Arizona — pointed to decreasing illegal immigration in the state. These estimates and claims rest on several annual efforts to count illegal immigrants in the U.S. The nonpartisan Pew Hispanic Center estimated that in 2008 the nationwide population was 11.9 million, and half a million in Arizona. The federal Department of Homeland Security and the Center for Immigration Studies, a Washington, D.C., research group that opposes increased immigration, agree on a figure of 10.8 million for 2009, with DHS putting the Arizona population at 460,000, down from 560,000 a year earlier. But as my print column notes this week, these estimates are limited by several factors that make it difficult for researchers to count this population. [REDACTED]. Thus estimates of the number of illegal immigrants in the country are indirect and possibly far off from the correct count. These studies rely on census surveys, and assume that about 10% of illegal immigrants aren’t counted

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 24788

in these surveys. But that figure largely is based on a 2001 survey of Mexican-born people living in Los Angeles. “I do not advise use of my estimated undercounts for the 2000 census outside of L.A. county, nor for migrants from other nations,” said study co-author Enrico Marcelli, assistant professor of sociology at San Diego State University. “However, demographers do not have any other empirical evidence at the moment with which to proceed.” One concern is that the nearly two in five households who didn’t respond to the 2001 survey may have included a disproportionately large number who also didn’t respond to census interviewers. Marcelli said further study would be needed to test that possibility, but he noted the extent of the efforts to select a representative sample and to put respondents at ease in order to elicit honest answers. “As far as I know, there has not been a new, serious attempt to estimate the undercount of illegal immigrants in the census,” said Steven Camarota, director of research for the Center for Immigration Studies. In 2005, Robert Justich, then a portfolio manager for Bear Stearns, co-authored a report suggesting the population of illegal immigrants “may be as high as 20 million people.” Jeffrey Passel, senior demographer for the Pew Hispanic Center, disputed that finding. For one thing, other data sources, such as U.S. birth rates and Mexico’s own census, don’t corroborate such a large number. If there were really so many more immigrants, than there would be more women of child-bearing age,

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 24889

and more births. And if instead the missing millions are mostly Mexican men working in the U.S. and sending money home, the flip side of that influx would be reflected as a gap in the Mexican census numbers. “Definitely the number is not as high as 20 million,” said Manuel Orozco, senior associate of the Inter- American Dialogue, a Washington, D.C., policy- analysis group. Justich, who now owns a music and film production firm, countered that immigrants from countries other than Mexico may make up the rest. However, he added that the number is no longer as high as 20 million. Larger estimates also sometimes are based on border-patrol counts of apprehensions, which are far from reliable proxies. No one is sure of how many people are missed for each one who is caught trying to cross into the U.S. illegally. Many of those who do get through may return quickly, or cross back and forth. Also, some people are caught more than once, inflating the count. “It seems like we’re not missing that many bodies in the United States,” said Camarota, referring to the gap between the 20 million figure and his own. The immigrant counters generally have seen a decline in the illegal-immigration population. “Economic drivers are very, very powerful” in lowering the illegal-immigrant population, said Hans Johnson, associate director of the Public Policy Institute of California. Others point to stepped-up enforcement efforts.

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 24990

However, because of all the assumptions baked into these numbers, such drops come with so much statistical uncertainty that they may not be statistically significant. “The methodology for doing these estimates is not really designed to measure year-to-year change,” Passel said. One key difference between his count and the federal agency’s: Homeland Security uses the Census Bureau’s American Community Survey, which has a much larger sample size than the Current Population Survey, which Passel used. “I developed all of my methodology and all of the things that go with it when there wasn’t an ACS,” Passel said, “and I haven’t gotten around to shifting to the new survey.” The ACS was introduced after the 2000 census, and may help overcome a problem with census numbers exposed in the last decennial census. [REDACTED] Census officials think these estimates have improved since 2000 thanks to the annual ACS surveys of three million households. ‘‘That’s the source we’re using to estimate the movement’’ of the foreign-born population, said Howard Hogan, the Census Bureau’s associate director for demographic programs. “It’s a huge improvement over anything we had available in the ’90s.” Still, the Census Bureau doesn’t ask people about their immigration status, in part because such questions may drive down overall response rates. Robert M. Groves, director of the Census Bureau, said he’d like to test that hypothesis. “We’re sort of data geeks here,” Groves said. “What we’d like to do to answer that question is an experiment.”

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 25091

That doesn’t mean that census interviewers don’t try to find and enumerate illegal immigrants. Groves compares counting that group to efforts to track another population that is hard to count, though not necessarily because of willful avoidance: people who are homeless. Census interviewers spend three days visiting soup kitchens, shelters and outdoor gathering spots such as under certain highway overpasses in Los Angeles. “You don’t have to look at that operation very long to realize that though it’s a heroic effort, there are all sorts of holes in it,” Groves said. As a result, the Census Bureau includes anyone counted in that effort in the overall population, but doesn’t break out a separate estimate of homeless people. “We would like to do estimates that have the smallest number of assumptions we can’t test,” Groves said. When it comes to counting illegal immigrants, “there are a set of assumptions that we know we can’t test. When we find ourselves in that situation, then we’re uncomfortable giving a Census Bureau estimate that is subject to all of those debates.” Further reading: Passel outlined methods for counting the illegal-immigrant population, while this paper analyzed some difficulties with the estimates. Earlier the Christian Science Monitor and I have examined these numbers. Immigration statistics have become a subject of debate in the U.K., as well.

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 25192

PX-057 From: Wilbur Ross [/O=EXCHANGELABS/ OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/ CN=6EA444C1E0EB42CFSDC621A7B6 D014B4-WLR] Sent: 9/19/2017 3:02:32 PM To: Davidson, Peter (Federal) [REDACTED]@doc.gov] Subject: Census Wendy and I spoke with the AG yesterday. Please follow up so we can resolve this issue today. WLR sent from my iPhone [STAMP] EXHIBIT Teramoto 10 8/24/18 TD

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 25293

PX-061

From: [REDACTED]@doc.gov [REDACTED] Sent: 9/18/2017 3:10:02 PM To: Gore, John (CRT) [REDACTED] Subject:Re: Call Hi. AG and Sec spoke. Pls let me know when you have a minute. Sent from my iPhone On Sep 16, 2017, at 3:55 PM, Gore, John (CRT) [REDACTED] wrote: Wendy: By this email, I introduce you to Danielle Cutrona from DOJ. Danielle is the person to connect with about the issue we discussed earlier this afternoon. Danielle: Wendy’s cell phone number is [REDACTED] Thanks. Sent from my iPhone On Sep 13, 2017, at 4:57 PM, Teramoto, Wendy (Federal) [REDACTED] wrote: Yes. CC’ing macie to set up. Look forward to connecting. W Sent from my iPhone On Sep 13, 2017, at 4:44 PM, Gore, John (CRT) [REDACTED] wrote Wendy:

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 25394

My name is John Gore, and I am an acting assistant attorney general in the Department of Justice. I would like to talk to you about a DOJ- DOC issue. Do you have any time on your schedule tomorrow (Thursday) or Friday for a call? Thanks. John M. Gore Acting Assistant Attorney General Civil Rights Division U.S. Department of Justice [REDACTED]

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 25495

PX-066

From: Cutrona, Danielle (OAG) [REDACTED] Sent: 9/18/2017 1:05:14 AM To: Teramoto, Wendy (Federal) [REDACTED] Subject: Re: Call Excellent. Thanks. Sent from my iPhone On Sep 17, 2017, at 8:25 PM, Teramoto, Wendy (Federal) [REDACTED] wrote: They connected. Thanks for the help. Wendy Sent from my iPhone On Sep 17, 2017, at 12:10 PM, Cutrona, Danielle (OAG) [REDACTED] wrote: Wendy, The Attorney General is available on his cell. His number is [REDACTED] He is in Seattle so he is 3 hours behind us. From what John told me, it sounds like we can do whatever you all need us to do and the delay was due to a miscommunication. The AG is eager to assist. Please let me know if you need anything else. You can reach me at [REDACTED] Thanks, Danielle Sent from my iPhone On Sep 17, 2017, at 10:08 AM, Cutrona, Danielle (OAG) [REDACTED] wrote:

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 25596

Checking now. Will let you know as soon as I hear from him. Sent from my iPhone On Sep 16, 2017, at 6:29 PM, Teramoto, Wendy (Federal) [REDACTED] wrote: Thanks. Danielle–pls let me know when the AG is available to speak to Secretary Ross. Thanks. Anytime on the weekend is fine too. W Sent from my iPhone On Sep 16, 2017, at 3:55 PM, Gore, John (CRT) [REDACTED] wrote: Wendy: By this email, I introduce you to Danielle Cutrona from DOJ. Danielle is the person to connect with about the issue we discussed earlier this afternoon. Danielle: Wendy’s cell phone number is [REDACTED] Thanks. [STAMP] EXHIBIT 26 WIT: Comstock DATE: 8/30/18 KLJ Sent from my iPhone On Sep 13, 2017, at 4:57 PM, Teramoto, Wendy (Federal) [REDACTED] wrote: Yes. CC’ing macie to set up. Look forward to connecting. W Sent from my iPhone

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 25697

On Sep 13, 2017, at 4:44 PM, Gore, John (CRT) [REDACTED] wrote: Wendy: My name is John Gore, and I am an acting assistant attorney general in the Department of Justice. I would like to talk to you about a DOJ-DOC issue. Do you have any time on your schedule tomorrow (Thursday) or Friday for a call? Thanks. John M. Gore Acting Assistant Attorney General Civil Rights Division U.S. Department of Justice [REDACTED]

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 25798

PX-003

From: John Maron Abowd (CENSUS/ADRM FED) [[email protected]] Sent: 2/20/2018 2:38:50 PM To: Phillips, John [REDACTED]; Ron S Jarmin (CENSUS/ADEP FED) [[email protected]] CC: Enrique Lamas (CENSUS/ADDP FED) [[email protected]]; Michael A Berning (CENSUS/ERD FED) [[email protected]]; Davies, Paul [REDACTED]; Hebert, Sybil [REDACTED]; Katherine Dodson Hancher (CENSUS/DEPDIR FED) [[email protected]] Subject: Re: Follow-up from our call Since you want Ron, please work through his assistant Kathy Hancher. Thanks, John John M. Abowd, PhD Associate Director and Chief Scientist Research and Methodology U.S. Census Bureau Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media From: Phillips, John [REDACTED] Sent: Tuesday, February 20, 2018 9:36:03 AM To: Ron S Jarmin (CENSUS/ADEP FED); John Maron Abowd (CENSUS/ADRM FED)

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 25899

Cc: Enrique Lamas (CENSUS/ADDP FED); Michael A Berning (CENSUS/ERD FED); Davies, Paul; Hebert, Sybil Subject: RE: Follow-up from our call Ron and John, Thank you for the responses below. After discussing internally, we have some additional questions. Can we schedule a call to discuss? Maybe keep it small for now (Ron, Enrique, John, Mike; John, Paul). Who should we work with on your staff for scheduling purposes? Sybil Hebert, copied here, handles my calendar. I look forward to speaking soon. Thanks. John From: John Maron Abowd (CENSUS/ADRM FED) [mailto:[email protected]] Sent: Thursday, February 15, 2018 8:51 AM To: Phillips, John [REDACTED] Cc: Enrique Lamas (CENSUS/ADDP FED) ; Michael A Berning (CENSUS/ERD FED) ; Davies, Paul [REDACTED]; Ron S Jarmin (CENSUS/ADEP FED) Subject: Re: Follow-up from our call Hi John, Ron asked me to coordinate the responses to your questions, and reply directly to you. Mike and Paul have also been in direct communication. Question 1: When does Census wish to begin including the Numident citizenship data in the

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 100259

PL94 reports? I gather from the write-up that the data must be released by April 1 of the year following a decennial census, which I presume would be April 1, 2021. The write-up also refers to annual CVAP tabulations each February based on 5-year ACS tabulations. How soon do you need approval/concurrence from us? Are you wanting to use Numident citizenship data in your February 2018 release? I doubt that we would be able to get this approved in time. Answer: The Census Bureau has not yet designated the vehicle of publication for this new product but has been asked to release a citizenship tabulation product by April 1st of 2021, alongside the P.L. 94-171 data. Question 2: For the purpose of our legal review, it would be helpful if you could point to specific authorities or specific projects within the SSA- Census Numident agreement that you believe support this activity. Can you please send us a project name/number and/or cite specific passages in the agreement? We will share this information with our lawyers to aid their review. Answer: Our impression was that the following paragraph in the authorized uses section of the current Numident agreement (Section 5.A.1b) includes the use of citizenship data for this project: b. The Census Bureau may use demographic characteristics in the Numident (such as age, sex, and race/Hispanic origin) missing from the Census Bureau’s administrative records as part

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 101260

of its decennial research program, and for current surveys, decennial census, survey, or other Title 13 administrative records for research and statistics uses. Specifically, the Census Bureau may use Numident data to improve decennial census operations including the un-duplication of public, private, and census lists; imputing missing data, and modeling to design and assign resources to carry out the census. [IMAGE] However, just to be sure we re-wrote that paragraph and submitted it to SSA OGC as a modification. That mod is still in SSA legal review. The revised paragraph is shown below: 1. Replace Section 5.A.1.b (quoted above) Authorized Uses of the Numident, paragraph one with the following revised paragraph: b. The Census Bureau may use demographic characteristics in the Numident (such as name, date of birth, age, sex, race/Hispanic origin, and country of origin) as the source for direct substi- tution or replacement to the Census Bureau’s decennial census research, testing and operations, and other demographic and economic survey improvements as well as for administrative records linked research and operations for statistical purposes. This usage includes the archiving and release of these directly substituted characteristics 72 years after the Census. Such authorized uses include the use of the Numident to improve decennial census operations including the un-duplication of public, private and census

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 102261

lists; imputing or directly substituting missing data, modeling to design and assign resources to carry out the census or surveys, and supporting administrative records research. Please let me know if you would like further clarification on either of these answers. Best, John John M. Abowd, PhD Associate Director and Chief Scientist Research and Methodology U.S. Census Bureau Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media From: Phillips, John [REDACTED] Sent: Wednesday, February 14, 2018 1:24:07 PM To: Ron S Jarmin (CENSUS/ADEP FED) Cc: Enrique Lamas (CENSUS/ADDP FED); John Maron Abowd (CENSUS/ADRM FED); Michael A Berning (CENSUS/ERD FED); Davies, Paul Subject: RE: Follow-up from our call Ron, Thank you for the write-up about enhancing the PL94/CVAP data with citizenship information from the Numident. As you mention, the lawyers will need to agree that this is covered under the SSA-Census Numident agreement. To help us

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 103262

with our internal reviews, can you please respond to the questions below? When does Census wish to begin including the Numident citizenship data in the PL94 reports? I gather from the writeup that the data must be released by April 1 of the year following a decennial census, which I presume would be April 1, 2021. The write-up also refers to annual CVAP tabulations each February based on 5-year ACS tabulations. How soon do you need approval/ concurrence from us? Are you wanting to use Numident citizenship data in your February 2018 release? I doubt that we would be able to get this approved in time. For the purpose of our legal review, it would be helpful if you could point to specific authorities or specific projects within the SSA-Census Numident agreement that you believe support this activity. Can you please send us a project name/number and/or cite specific passages in the agreement? We will share this information with our lawyers to aid their review. At some point, I suspect we will have additional questions about the proposal and may need more details about how the Numident citizenship data would be incorporated into the PL94/CVAP/2020 processes and how disclosure avoidance will be handled (in somewhat less technical terms). But for now, I think the questions above are more pressing. We are briefing our leadership on this subject and that too may generate more questions – I will keep updated. Thanks, John

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 104263

From: Ron S Jarmin (CENSUS/ADEP FED) [mailto:[email protected]] Sent: Wednesday, January 24, 2018 9:34 AM To: Phillips, John [REDACTED] Cc: Enrique Lamas (CENSUS/ADDP FED) ; John Maron Abowd (CENSUS/ADRM FED) ; Michael A Berning (CENSUS/ERD FED) Subject: Follow-up from our call John, Please see attached short description of a plan to enhance the CVAP data we provide to DOJ using citizenship information from the NUMIDENT which a team here as analyzed and found to fit for this purpose. Only block level tabulations will leave the Bureau and the CVAP microdata are NOT part of the Decennial record system that gets archived and made available to the public in 72 years. I believe the agreement we have with SSA that expires in 2019 covers this type of use, but lawyers will need to agree. Since we’ll need to extend that agreement, that can be an opportunity to ensure this type of use is covered. Please let know if you have any questions and let’s discuss how best to proceed. Thanks Ron Jarmin, PhD. Associate Director for Economic Programs, and Performing the Non-Exclusive Functions and Duties of the Director, U.S. Census Bureau Office 301.763.1858, [email protected] census.gov Connect with us on Social Media

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 105264

PX-071

From: [email protected] [[email protected]] Sent: 12/22/2017 8:36:41 PM To: Karen Kelley [PII] Subject: Fwd: Request to Reinstate Citizenship Question On 2020 Census Questionnaire FYI. Will let you know what I hear. Sent from my iPhone Begin forwarded message: From: “Ron S Jarmin (CENSUS/ADEP FED)” Date: December 22, 2017 at 3:32:12 PM EST To: [PII] Cc: “Enrique Lamas (CENSUS/ADDP FED)” Subject: Request to Reinstate Citizenship Question On 2020 Census Questionnaire Arthur, Thank you for your letter dated 12/12/2017 regarding improving the quality of citizenship information for DOJ enforcement of the Voting Rights Act. Let me start by saying the Bureau is fully supportive of providing DOJ with the highest quality statistical information possible. To that end , I directed staff to review all possible ways to address the needs expressed in the letter. They have now briefed me and their findings suggest that the best way to provide PL94 block-level data

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 106265

with citizen voting population by race and ethnicity would be through utilizing a linked file of administrative and survey data the Census Bureau already possesses. This would result in higher quality data produced at lower cost. I suggest we schedule a meeting of Census and DOJ technical experts to discuss the details of this proposal. We look forward to working with you on this important statistical matter. Happy Holidays Ron Jarmin, PhD. Associate Director for Economic Programs, and Performing the Non-Exclusive Functions and Duties of the Director U.S. Census Bureau Office 301.763.1858, [email protected] census.gov Connect with us on Social Media [STAMP] EXHIBIT 7 WIT: JARMIN DATE: 8/20/18 KLJ

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 107266

PX-75 (R)

From: Ron S Jarmin (CENSUS/ADEP FED) [[email protected]) Sent: 2/6/2018 8:42:03 PM To: Kelley, Karen (Federal) [REDACTED] CC: Lamas, Enrique [enrique.lamas@census. gov] Subject: DOJ Karen, I spoke with Art Gary. He has spoken with DOJ leadership. They believe the letter requesting citizenship be added to the 2020 Census fully describes their request. They do not want to meet. Thanks Ron Sent from my iPhone

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 108267

Interim pre-decisional Template sent by email [WATERMARK] DRAFT January XX, 2018 Senator Harris, et al. Thank you for your January 5, 2018, expressing concern regarding the Department of Justice’s request to add a citizenship question to the 2020 Census questionnaire. The U.S. Census Bureau has a well-established process for considering requests for new questions to the Decennial Census and the American Community Survey. The requested data must fulfill legal and regulatory requirements established by the Congress, and the Census Bureau works with the Office of Management and Budget (OMB) to review and assess the justification of the new content or question. While the discretionary authorization for defining new content or questions resides with the Secretary of Commerce, the 5-step process below is employed by the Census Bureau and OMB to make a determination. Once each of these steps are completed, a new question can be added to the 2020 Census. Step One: With the exception of technical questions needed to collect accurate data, all questions on the various census forms generate data in response to request for the Congress or other agencies in the Executive Branch.

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 109268

Step Two: Upon determining that a new question is warranted, the Census Bureau must notify the Congress of its intent to add the question. As you may know, by law, the Census Bureau notified the Congress of the topics to be covered in the 2020 Census on March 31, 2017, and must deliver the specific questions by March 31, 2018. Step Three: The Census Bureau then must notify the public, and invite comments regarding the change in the questionnaire with a Federal Register Notice. Step Four: The Census Bureau must test the wording of the new question. Step Five: The Census Bureau must make additional operational adjustments, beyond testing, to include new content. This includes redesigning each data capture method as well as training modules for enumerators. We will keep you apprised of any developments regarding the citizenship question. If you have any additional questions or would like to discuss the formal process in detail, please have a member of your staff contact the Census Bureau’s Office of Congressional and Intergovernmental Affairs on (301) 763-6100. Sincerely,

Wilbur L. Ross, Jr. Secretary Department of Commerce

78228 • ACLU • APPENDIX 03 (DOCS 10 THRU 19) AL 3/22/19 110269

CONSTITUTIONAL ACCOUNTABILITY CENTER [LETTERHEAD] ______

March 22, 2018 The Honorable Secretary Wilbur Ross U.S. Department of Commerce 1401 Constitution Avenue NW Washington, DC 20230 Dear Secretary Ross: We the undersigned legal organizations write to urge you to reject the Department of Justice’s request that you add a mandatory question to the 2020 Census asking all persons to divulge their citizenship status. A new, untested citizenship question would be an end-run around the Constitution’s text, history, and values. It cannot be squared with the federal government’s constitutional obligation to ensure a national count of all persons-regardless of where they are from or their immigration status. Our Constitution establishes a democracy premised on the idea that all persons deserve equal representation in our government. To ensure a proper count of the nation’s population and a proper apportionment of representatives, the Constitution explicitly requires an “actual Enumeration” of the people, imposing on the federal government the duty to count the “whole number of persons in each State.”1 This critical

1 U.S. Const. art. I,§ 2, cl. 3; amend. XIV,§ 2.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 111270

constitutional language imposes a clear duty on the federal government: it must count all people living in the United States, whether they are citizens or non-citizens, whether they were born in the United States or in a distant part of the world. As the Constitution’s text and history dictate, the Constitution requires the federal government to count “the whole body of the people” without exception.2 lt draws no distinction between citizens and non-citizens, but rather requires that the “whole immigrant population should be numbered with the people and counted as part of them.”3 Adding the new citizenship question proposed by the Department of Justice would undermine the Census Bureau’s constitutional commitment to count all persons. It would also result in inaccurate data, thereby biasing congressional apportionment, redistricting, and funding decisions, for an entire decade, and producing harmful inequalities which would last even longer. Overwhelming evidence shows that this new question, if it becomes a part of the 2020 Census, will deter participation by immigrants across the country, who do not want an official record of their immigration status and fear that their responses will be used by the government to harm them and their families. The Census Bureau’s own data demonstrates “an unprecedented groundswell in confidentiality and data sharing concerns, particularly among immigrants or those who live with immigrants.”4 In the run up to the 2020

2 Cong. Globe, 39th Cong., 1st Sess. 385 (1866). 3 Id. at 432.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 112271

Census, “researchers heard respondents express new concerns about topics like the ‘Muslim Ban,’ discomfort ‘registering’ other household members by reporting their demographic characteristics, the dissolution of the ‘DACA’ ... program, [and] repeated references to Immigration and Customs Enforcement.”5 Adding a citizenship question to the 2020 Census-given the overwhelming evidence that it will chill participation and produce inaccurate responses—would break faith with the Constitu- tion’s mandate for a head count of the entire nation. Although the Department of Justice urges the addition of a citizenship question to the 2020 Census, it offers no reason to doubt what the latest Census Bureau data shows: asking all persons to divulge their citizenship status will chill participation by noncitizens and citizens alike and produce inaccurate data. Instead, the DOJ maintains that a new citizenship question will ensure better enforcement of the Voting Rights Act. This is false. Since the passage of the Voting Rights Act in 1965, the Census has never asked all persons to report their citizenship. In other words, a mandatory question on citizenship has never been necessary to ensure robust protection of the

4 Mikelyn Meyers, U.S. Census Bureau, Respondent Confidentiality Concerns and Possible Effects on Response Rates and Data Quality for the 2020 Census 15 (Nov. 2, 2017), https://www2.census.gov/cac/nac/meetings/2017-11/ Meyers-NAC-Confidentiality-Presentation.pdf. 5 Memorandum from Ctr. for Survey Measurement to the Assoc. Directorate for Research & Methodology, Re: Respondent Confidentiality Concerns 1 (Sept. 20, 2017), https://www2.census.gov/cac/nac/meetings/2017-11/Memo- Regarding-Respondent-Confidentiality-Concerns.pdf.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 113272

right to vote. That is just as true now as it was in 1965 when the Voting Rights Act was passed. The Justice Department’s effort to game the Census and manipulate the national head count our Framers wrote into the Constitution should be rejected. Failing to count all persons in the United States, as the Constitution mandates, would deal a huge blow to our democracy. The stakes are high, and there are no do-overs permitted—we must get it right, and get it right now. Sincerely, Asian Americans Advancing Justice— AAJC Campaign Legal Center Constitutional Accountability Center Democracy Forward Lawyers’ Committee for Civil Rights Under Law NAACP Legal Defense and Education Fund, Inc. United To Protect Democracy Voting Rights Institute

cc: Donald F. McGahn, White House Counsel Michael J. Walsh, Jr., Deputy General Counsel, Department of Commerce Hon. Ron Johnson, Chairman, Senate Homeland Security and Governmental Affairs Committee Hon. Claire McCaskill Ranking Member, Senate Homeland Security and Governmental Affairs Committee Hon. Trey Gowdy Chairman, House Committee on Oversight and Government Reform Hon. Elijah Cummings Ranking Member, House Committee on Oversight and Government Reform

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 114273

PX-085

From: Langdon, David (Federal) [REDACTED] Sent: 5/24/2017 10:51:56 PM To: Blumerman, Lisa M [[email protected]] Subject: Fwd: Requested Information – Legal Review All Residents... Fyi on the citizenship question below. Can you provide a short answer? Ideally this evening. ------Original message ------From: “Langdon, David (Federal)” Date: 05/24/2017 5:53 PM (GMT-05:00) To: “Reist, Burton H (CENSUS/ADDC FED)” Cc: “Creech, Melissa L”, “Dinwiddie, James L” Subject: RE: Requested Information – Legal Review All Residents... Actually, the Secretary seemed interested on subjects and puzzled why citizenship is not included in 2020. [REDACTED] Say, citizenship. What criteria drives us to put it on ACS but not 2020?

From: Reist, Burton H (CENSUS/ADDC FED) [mailto:[email protected]] Sent: Wednesday, May 24, 2017 5:42 PM To: Langdon, David (Federal) [REDACTED]

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 115274

Cc: Creech, Melissa L ; Dinwiddie, James L Subject:Re: Requested Information – Legal Review All Residents... David, Melissa and I will be in early tomorrow. If you need anything let us know. Lisa and I are also happy to discuss the Lifecycle stuff I sent and answer the questions you have. Burton

From: Langdon, David (Federal) Sent: Wednesday, May 24, 2017 5:24:30 PM To: Reist, Burton H (CENSUS/ADDC FED) Cc: Melissa L Creech (CENSUS/PCO FED); James L Dinwiddie (CENSUS/ADDC FED) Subject: RE: Requested Information – Legal Review All Residents... Thank you! I apologize for not answering sooner, but I honestly have been in meeting with SWR all afternoon. (Not the norm.) This is a lot to digest, but [REDACTED]

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 116275

From: Reist, Burton H (CENSUS/ADDC FED) [mailto: [email protected]] Sent: Wednesday, May 24, 2017 4:10 PM To: Langdon, David (Federal) [REDACTED] Cc: Creech, Melissa L ; Dinwiddie, James L Subject: Fw: Requested Information – Legal Review All Residents... This is the more complete set of documents that I referenced in my earlier email. Burton

From: Misty L Reed (CENSUS/DEPDIR FED) Sent: Wednesday, May 24, 2017 4:02 PM To: Reist, Burton H (CENSUS/ADDC FED) Subject: Requested Information – Legal Review All Residents... Hotspots are amazing and luckily I scanned the files (Melissa gave me hard copies). Let me know if there’s anything else I can provide. Thanks, Misty Misty Reed, PhD, PMP, Special Assistant, Communications Directorate, U.S. Census Bureau Office 301.763.0228 Cell [REDACTED] [email protected] census.gov Connect with us on Social Media

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 117276

PX-088

From: Comstock, Earl (Federal) [REDACTED] Sent: 5/2/2017 2:19:11 PM To: Wilbur Ross [REDACTED] CC: Herbst, Ellen (Federal) [REDACTED] Subject: Re:Census I agree Mr Secretary. On the citizenship question we will get that in place. The broad topics were what were sent to Congress earlier this year as required. It is next March — in 2018 – when the final 2020 decennial Census questions are submitted to Congress. We need to work with Justice to get them to request that citizenship be added back as a census question, and we have the court cases to illustrate that DoJ has a legitimate need for the question to be included. I will arrange a meeting with DoJ staff this week to discuss. Earl Sent from my iPhone >On May 2, 2017, at 10:04 AM, Wilbur Ross [REDACTED] wrote: [REDACTED] Worst of all they emphasize that they have settled with congress on the questions to be asked. I am mystified why nothing have been done in response to my months old request that we include the citizenship question. Why not? [REDACTED] >Sent from my iPhone

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 118277

Talking Points • The Census Bureau today received a letter from the Department of Justice requesting that a question on citizenship be added to the 2020 Census. • The Census Bureau follows a well-established process when adding questions to the decennial census based on the recognition that the data must fulfill legal and regulatory requirements established by the Congress. While the discretionary authority for defining the questions on either the American Community Survey or the Decennial Census questionnaire resides with the Secretary of Commerce, the Census Bureau works with the Office of Man- agement and Budget (OMB) to review and assess the justification of the new question. • By law, the Census Bureau must provide the questions for the 2020 Census and American Community Survey to Congress by March 31 – two years prior to taking the national headcount. • Historically, a citizenship question was asked periodically over the history of the census, and most recently from 1980 to 2000 as part of a decennial census long form questionnaire that provided socio-economic and housing characteristic data of the population. • The Census Bureau currently asks citizenship on its nationwide American Community Survey, a survey conducted nationwide every year among 3.5 million addresses.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 119278

Response to Query “The Census Bureau today received a letter from the Department of Justice requesting that a question on citizenship be added to the 2020 Census. The Census Bureau follows a well-established process when adding questions to the decennial census based on the recognition that the data must fulfill legal and regulatory requirements established by the Congress. While the discretionary authority for defining the questions on either the American Community Survey or the Decennial Census questionnaire resides with the Secretary of Commerce, the Census Bureau works with the Office of Management and Budget (OMB) to review and assess the justification of the new question. The Census Bureau does ask citizenship on its nationwide American Community Survey, a survey conducted nationwide every year among 3.5 million addresses.” Process of Adding Content to the Census/Survey Questionnaire • Step One – With the exception of operational questions needed to collect accurate data, all questions on the various census questionnaires generate data in response to requests from the Congress or other agencies in the Executive Branch. Upon receiving a request lawyers at the Department of Commerce work closely with the Census Bureau staff to determine whether the data fulfill legal, regulatory or Constitutional requirements. Within this process, the Census Bureau also consults with the OMB.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 120279

• Step Two – Upon determining that a new question is warranted, the Census Bureau notifies Congress of its intent to add the question through its submission of the proposed questions for the 2020 Census. By law, the Census Bureau notified the Congress of the subjects to be covered by the 2020 Census on March 28, 2017. The Census Bureau must deliver the specific questions by March 31, 2018. This is an intentional process designed to give the Congress the ability to review the subjects and questions on the questionnaire before they are finalized. • Step Three – The Census Bureau must test the wording of the new question. It is too late to add a question to the 2018 End-to-End Census Test, so additional testing on a smaller scale would need to be developed and implemented as soon as possible. This test would also require approval from OMB, which includes notifying the public and inviting comments through a Federal Register Notice (FRN). The updated FRN needs to be cleared by OMB prior to a new 30-day FRN posting. The Census Bureau must respond to comments from the public after 30 days. Then OMB can issue final approval. • Step Four – The Census Bureau must make additional operational adjustments, beyond testing, to include new content. This includes re-designing the paper questionnaires and adjusting the paper data capture system. For all automated data collection instruments (including Internet self-response, Census

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 121280

Questionnaire Assistance, and Nonresponse Followup), the additional question will require system redevelopment, for English and all supported non-English languages. In addition, the training for the enumerators and Census Questionnaire Assistance agents will need redevelopment. • Step Five – Based on the result of the testing, the Census Bureau must finalize the actual 2020 Census questionnaires (paper and automated). The Census Bureau then must submit for OMB approval of the 2020 Census information collection. This submission also requires notifying the public and inviting comments through a Federal Register Notice (FRN), as detailed in Step 3.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 122281

PX-095

From: Wilbur Ross [REDACTED] Sent: 8/10/2017 7:38:25 PM To: Comstock, Earl (Federal) [REDACTED] Subject: Re: Census Matter I would like to be briefed on Friday by phone. I probably will need an hour or so to study the memo first. [REDACTED] WLR sent from my iPad >On Aug 9, 2017, at 10:24 AM, Comstock, Earl (Federal) <[REDACTED]> wrote:> > >PREDECISIONAL AND ATTORNEY-CLIENT PRIVILEGED > >Mr. Secretary – we are preparing a memo and full briefing for you on the citizenship question. The memo will be ready by Friday, and we can do the briefing whenever you are back in the office. [REDACTED] > >Earl > >On 8/8/17, 1:20 PM, “Wilbur Ross” [REDACTED] wrote: >[REDACTED] Were you on the call this morning about Census? [REDACTED] where is the DoJ in their analysis? If they still have not come to a

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 123282

conclusion please let me know your contact person and I will call the AG. Wilbur Ross > >Sent from my iPhone >>On Aug 8, 2017, at 10:52 AM, Comstock, Earl (Federal) [REDACTED] wrote: >> >>[REDACTED] > >

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 124283

PX-097

To: Wilbur Ross [REDACTED] From: Comstock, Earl (Federal) Sent: Tue 8/8/2017 7:44:29 PM Importance: Normal Subject: Re: [REDACTED] Received: Tue 8/8/2017 7:44:29 PM [REDACTED PARAGRAPH] Will be back shortly with an update on the census question. I have two attorneys in the DoC General Counsel’s office working on it. Earl On 8/8/17, 1:20 PM, “Wilbur Ross” [REDACTED] wrote: >[REDACTED] Were you on the call this morning about Census? [REDACTED] where is the DoJ in their analysis? If they still have not come to a conclusion please let me know your contact person and I will call the AG. Wilbur Ross Sent from my iPhone >On Aug 8, 2017, at 10:52 AM, Comstock, Earl (Federal) [REDACTED] wrote: > >[REDACTED]

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 125284

PX-098

PRE-DECISIONAL

Submission of the 2020 Census and American Community Survey Questions to Congress

Briefing for the XXXX

February##, 2018

( United Slates· IU.S. Oepanment of Commerce ensus ~:r~~,..... ,,.n .. ,., ... l(IO, _ ....,..., ,, .. ,.."'""'

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 126285

PRE-DECISIONAL

2020 Census and American Community Survey Subjects and Questions Requirements

Section 141(f) of the Census Act requires that the ~o1>o

✓ This document wa s iss ued on March 28, 2017.

The Census Act also requires that the questions included in the next census be submitted to Congress no later than 2 years before the census date.

► A document that meets this requirement for the 2020 Cen sus and the ACS will be submitted to Congre ss by March 31, 2018.

The Census Bureau is required by Section 141(f) of the Census Act to submit the subjects for the next census to Congress no later than 3 years before April 1, 2020. We delivered the subjects in March 2017. The Census Act also requires the questions included in the next census be submitted to Congress no later than 2 years before April 1, 2020. A document that meets this requirement for the 2020 Census and the ACS will be submitted to Congress by March 31, 2018.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 127286

PRE-DECISIONAL How a Question Becomes Part of the Census {short form) Steps in the Process

Authority • The discretionary authority for defining the questions on the Decennial Census Short Form resides with the Secretary of Commerce.

Review of Request • Requests undergo legal, technical, and policy review to determine whether the data fulfill lega l, regulatory, or Constitutional req uirements.

Notification • Upon determining a new question is warranted, the Census Bureau must notify Congress of its intent to add the question. • The Census Bureau also solicits public comment through a Federal Register Notice.

Testing • If the question is not currently used in an ongoing survey, the Census Bureau must test the wording of the new question.

Operational Adjustments • The Cen sus Bureau must make operational adjustment s to all dat a collection and processing systems to include the approved, new question.

( United Stales· I U.S. Dcpanment of Commerce ~r=•~~>l1t,Ml1KM"1•hnn ___ensus ...,,..., rMSm-9<"1

The Census Bureau follows a well-established process when adding questions to the decennial census based on the recognition that the data must fulfill legal and regulatory requirements established by the Congress. The discretionary authority for defining the questions on the Decennial Census Short Form resides with the Secretary of Commerce. Requests undergo legal, technical, and policy review to determine whether the data fulfill legal, regulatory, or Constitutional requirements. Upon determining a new question is warranted, the Census Bureau must notify Congress of its intent to add the question. By law, the Census Bureau notified the Congress of the topics to be covered by the 2020 Census on March 31, 2017.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 128287

The Census Bureau must deliver the specific questions by March 31, 2018. The Census Bureau also notifies the public and invites comments regarding the change in the questionnaire with a Federal Register Notice (FRN). The Census Bureau must test the wording of the new question. If approved, the Census Bureau must make operational adjustments to all data collection and processing systems to include the new question. *********************************************** Requests undergo legal review of the justification by DOC; technical review by the Census Bureau; and policy review by DOC and OMB. More detail if pressed further: The Census Bureau follows a well-established process when adding questions to the decennial census based on the recognition that the data must fulfill legal and regulatory requirements established by the Congress. While the discretionary authority for defining the questions on either the American Community Survey or the Decennial Census Short Form resides with the Secretary of Commerce, the Census Bureau works with the Office of Management and Budget (OMB) to review and assess the justification of the new content or question. Step One – With the exception of technical questions needed to collect accurate data, all questions on the various census forms generate

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 129288

data in response to requests from the Congress or other agencies in the Executive Branch. Upon receiving a request lawyers at the Department of Commerce work closely with OMB to determine whether the data fulfill legal, regulatory or Constitutional requirements. Step Two – Upon determining that a new question is warranted, the Census Bureau must notify Congress of its intent to add the question. This is particularly important for the 2020 Census Ques- tionnaire. By law, the Census Bureau notified the Congress of the topics to be covered by the 2020 Census on March 31, 2017. The Census Bureau must deliver the specific questions by March 31, 2018. This is an intentionally process designed to give the Congress the ability to review the topics and questions on the questionnaire before they are finalized. If an additional topic is required, it is imperative that Congress be notified as soon as possible. Step Three – The Census Bureau then must notify the public, and invite comments regarding the change in the questionnaire with a Federal Register Notice (FRN). The updated FRN needs to be cleared by OMB prior to a new 30-day FRN posting. The Census Bureau must respond to comments from the public after 30 days. Then OMB can issue final approval. Step Four – The Census Bureau must test the wording of the new question. With respect to a possible question on citizenship, no testing would be required because we would use a question identical to what we already use in the ACS.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 130289

Step Five – The Census Bureau must make additional operational adjustments, beyond testing, to include new content. This includes re-designing the paper questionnaire and adjusting the paper data capture system. For Internet self-response, the additional question will require system redevelopment, once for English and then again for Spanish. The Census Questionnaire Assistance operation will require development as well. Finally, the Nonresponse Followup data collection instruments will need to be redesigned, and the training modules for the enumerators will need further development. Once each of these steps are completed a new question can be added to the 2020 Census.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 131290

PX-100

UNITED STATES DEPARTMENT OF COMMERCE Economics and Statistics Administration U.S. Census Bureau Washington, DC 20233-0001 [LOGO] January 3, 2018 MEMORANDUM FOR: Ron S. Jarmin Performing the Non- exclusive Functions and Duties of the Director From: John M. Abowd [signature] Chief Scientist and Associate Director for Research & Methodology Prepared by: Stephen Buckner, Patrick Cantwell, Joanne Crane, Melissa Creech, Burton Reist, and James Whitehorne Subject: Summary of Quality/Cost of Alternatives for Meeting Department of Justice Request for Citizenship Data The Department of Justice has requested census block-level citizen voting-age population estimates by OMB-approved race and ethnicity categories from the 2020 Census of Population and Housing. These estimates are currently provided in two related data products: the PL94-171 redistricting data (PL94),

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 132291

produced by April 1st of the year following a decennial census under the authority of 13 U.S.C. Section 141, and the Citizen Voting Age Population by Race and Ethnicity (CVAP) tables produced every February from the most recent five-year American Community Survey data. The PL94-171 data are released at the census block level. The CVAP data are released at the census block group level. There are three alternatives for meeting the DoJ request: Alternative A: Maintain the status quo for data collection, preparation and publication. After the regular PL94 and CVAP data have been published in 2021, prepare a special product for DoJ that combines these tables to produce a Census Bureau best estimate of the block-level citizen voting age population by race and ethnicity. This would be similar to the approach now used to support the Section 203 requirements of the Voting Rights Act. Alternative B: Add a citizenship question to the 2020 Census questionnaire. Process the citizenship question using the 2020 Census data processing system, including using administrative records where currently authorized and implemented. Produce the block-level tables of citizen voting age population by race and ethnicity during the publication phase of the 2020 Census. Alternative C: Do not add a citizenship question to the 2020 Census questionnaire. Add the capability to link an accurate, edited citizenship variable from administrative records to the final 2020 Census microdata files. Produce the block-level tables of citizen voting age population by race and ethnicity

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 133292

during the publication phase of the 2020 Census using the enhanced 2020 Census microdata. An analysis of the cost and quality implications of each alternative follows. [CENSUS BUREAU LOGO AND URL] For Alternative A, the cost would be similar to the cost of the every-five-year production of language determinations to support DoJ enforcement of Section 203 of the Voting Rights Act. We estimate that the incremental cost of producing Alternative A is approximately $200,000. This estimate is based on the total cost of producing the 2016 language determinations. These costs are the salaried time of staff in the Redistricting Office, the Policy Coordination Office, the American Community Survey Office, and the Center for Statistical Research and Methodology. We estimate that delivering block-level citizen voting age population tables by race and ethnicity based on statistical modeling of the 2020 PL94-171 and CVAP tables would be a similar effort in terms of both cost and quality. For Alternative B, our estimate of the incremental cost is as follows. We estimate that there will be approximately 145 million addresses in the 2020 Census. Recent American Community Survey (ACS) data indicate that 9.8% of households contain at least one non-citizen. This yields an estimate of 14.2 million households with at least one non-citizen. Our analysis of 2010 response data for both the American Community Survey and the Short-Form Census indicates that the presence of a question on citizenship suppressed response by 5.1 percentage points for these households. Based on our working assumption of a response rate of 60.5% in the 2020 census, a

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 134293

reduction of 5.1 percentage points for households containing at least one non-citizen will increase the Nonresponse Follow-up (NRFU) workload by approximately 700,000 households, or approximately 0.5 percentage points. We estimate that for each percentage point increase in NRFU the cost of the 2020 Census increases by approximately $55 million. Accordingly, the addition of a question on citizenship could increase the cost of the 2020 Census by at least $27.5 million. It is worth stressing that this cost estimate is a lower bound. Our estimate of $55 million for each percentage point increase in NRFU is based on an average of three visits per household. We expect that we would make a total of six visits to households containing non-citizens who did not self- respond. We also believe that the decrease in response for these households in 2020 could be greater than the 5.1 percentage points we observed during the 2010 Census. Alternative B would most likely deliver higher quality block-level citizen voting age population by race and ethnicity data than Alternative A because it is based on obtaining a direct report of citizenship status for each respondent. But it would result in lower quality enumeration data. Because of the estimated 5.1 percentage point increase in NRFU for households with noncitizens and because NRFU is less accurate than self-reports, there is a decrease the coverage quality of the census. We estimate that asking the citizenship question would result in 154,000 fewer correct enumerations. This is also a lower bound estimate on the loss of accuracy. For Alternative C, the cost estimate has not yet been fully vetted. The estimated cost is less than

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 135294

$1,000,000. In the current system of administrative data acquisition and processing for the 2020 Census, there are ten remaining data acquisitions between January 1, 2018 and April 1, 2020. We include the cost of these data acquisitions in our incremental cost estimate for Alternative C because it would no longer be an option to discontinue the acquisition of these data and use only the already acquired files. In addition, we include the cost of two senior analysts. The analysts would do the required integration of the edited citizenship into the 2020 Census microdata. This estimate does not include the cost of modifying existing Memoranda of Understanding with the agencies that supply the data used to acquire the citizenship variable because those negotiations are already in progress, and would continue. This estimate also does not include the cost of negotiating a new MOU with the United States Citizen and Immigration Services to acquire those data, refreshed appropriately over the 2020 Census life cycle because we have not yet gathered the required information from USCIS to estimate this cost. Alternative C delivers higher quality data than Alternative B for DoJ’s stated uses. Our primary data sources for the administrative record citizen variable require proof of citizenship. For this reason, they are very accurate. There is good evidence that citizenship is accurately reported by citizens, but less accurately self-reported by household responders for noncitizens. This accuracy deficit in the self- responses may be due to the inherent difficulty of the respondent knowing the citizenship status of everyone in the household. Proxy respondents are even less likely to know the citizenship status of all members of the household. It may also be due to the sensitivity of the

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 136295

citizenship question itself. For DoJ’s stated purposes, Alternative C produces the highest quality data. Alternative A is not very costly and does not harm the quality of the census count. Alternative B better addresses DoJ’s stated uses; however, it is very costly and does harm the quality of the census count by increasing erroneous enumerations. Alternative C even better meets DoJ’s stated uses, is comparatively far less costly than Alternative B, and does not harm the quality of the census count. For these reasons, we recommend Alternative C for meeting the Department of Justice data request.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 137296

PX-101

From: Ron S Jarmin (CENSUS/ADEP FED) [[email protected]] Sent: 1/3/2018 6:45:55 PM To: Gary, Arthur (JMD) [REDACTED] CC: Enrique Lamas (CENSUS/ADDP FED) [[email protected]] Subject: Re: Request to Reinstate Citizenship Question On 2020 Census Questionnaire Gary, I’m bringing technical, program and legal folks. It would be good if some technical folks on the DOJ side were there so we can ensure we understand and can meet your requirements. Thursday and Friday are the most open for us, but we're flexible and can shuffle to meet earlier in the week if that’s preferable. Thanks Ron Jarmin, PhD. Associate Director for Economic Programs, and Performing the Non-Exclusive Functions and Duties of the Director U.S. Census Bureau Office 301.763.1858, [email protected] census.gov Connect with us on Social Media

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 138297

From: Gary, Arthur (JMD) [REDACTED] Sent: Tuesday, January 2, 2018 2:21:05 PM To: Ron S Jarmin (CENSUS/ADEP FED) Cc: Enrique Lamas (CENSUS/ADDP FED) Subject:RE: Request to Reinstate Citizenship Question On 2020 Census Questionnaire It should work fine — let me get back to you. Best wishes to you for 2018 as well. Thanks, Art Arthur E. Gary General Counsel Justice Management Division U.S. Department of Justice Two Constitution Square, Suite 8E.500 145 N. Street, NE Washington, DC 20530 202-514-3452 (OGC main line) NOTICE: This email (including any attachments) is intended for the use of the individual or entity to which it is addressed. It may contain information that is privileged, confidential, or otherwise protected by applicable law. If you are not the intended recipient (or the recipient’s agent), you are hereby notified that any dissemination, distribution, copying, or use of this email or its contents is strictly prohibited. If you received this email in error, please notify the sender immediately and destroy all copies.

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 139298

From: Ron S Jarmin (CENSUS/ADEP FED) [mailto:[email protected]] Sent: Tuesday, January 02, 2018 1:59 PM To: Gary, Arthur (JMD) <[REDACTED] Cc: Enrique Lamas (CENSUS/ADDP FED) Subject: Re: Request to Reinstate Citizenship Question On 2020 Census Questionnaire Arthur, Happy New Year! Would the late next week work for a meeting? Best Ron Jarmin, PhD. Associate Director for Economic Programs, and Performing the Non-Exclusive Functions and Duties of the Director U.S. Census Bureau Office 301.763.1858, [email protected] census.gov Connect with us on Social Media

From: Gary, Arthur (JMD) [REDACTED] Sent: Friday, December 22, 2017 4:16:35 PM To: Ron S Jarmin (CENSUS/ADEP FED) Cc: Enrique Lamas (CENSUS/ADDP FED) Subject: RE: Request to Reinstate Citizenship Question On 2020 Census Questionnaire Dr. Jarmin – thank you for your response. We look forward to meeting with you and your team in early January. Best regards

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 140299

Arthur E. Gary General Counsel Justice Management Division U.S. Department of Justice Two Constitution Square, Suite 8E.500 145 N. Street, NE Washington, DC 20530 202-514-3452 (OGC main line) NOTICE: This email (including any attachments) is intended for the use of the individual or entity to which it is addressed. It may contain information that is privileged, confidential, or otherwise protected by applicable law. If you are not the intended recipient (or the recipient’s agent), you are hereby notified that any dissemination, distribution, copying, or use of this email or its contents is strictly prohibited. If you received this email in error, please notify the sender immediately and destroy all copies.

From: Ron S Jarmin (CENSUS/ADEP FED) [mailto:[email protected]] Sent: Friday, December 22, 2017 3:32 PM To: Gary, Arthur (JMD) <[REDACTED] Cc: Enrique Lamas (CENSUS/ADDP FED) Subject: Request to Reinstate Citizenship Question On 2020 Census Questionnaire Arthur, Thank you for your letter dated 12/12/2017 regarding improving the quality of citizenship information for DOJ enforcement of the Voting Rights Act. Let me start by saying the Bureau is

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 141300

fully supportive of providing DOJ with the highest quality statistical information possible. To that end, I directed staff to review all possible ways to address the needs expressed in the letter. They have now briefed me and their findings suggest that the best way to provide PL94 block-level data with citizen voting population by race and ethnicity would be through utilizing a linked file of administrative and survey data the Census Bureau already possesses. This would result in higher quality data produced at lower cost. I suggest we schedule a meeting of Census and DOJ technical experts to discuss the details of this proposal. We look forward to working with you on this important statistical matter. Happy Holidays Ron Jarmin, PhD. Associate Director for Economic Programs, and Performing the Non-Exclusive Functions and Duties of the Director U.S. Census Bureau Office 301.763.1858, [email protected] census.gov Connect with us on Social Media

78228 • ACLU • APPENDIX 04 (DOCUMENTS 20 THRU 28) AL 3/22/19 142301

PX-102

*** INTERNAL CENSUS USE ONLY***

Alternative Sources of Citizenship Data for the 2020 Census Prepared for: John M. Abowd Through: John L. Eltinge Prepared by: Michael Berning, J. David Brown, Misty Heggeness, Shawn Klimek, Lawrence Warren, and Moises Yi Disclosure review: Amy Lauger (CBDRB-2017-CDAR-001) The statistics in this report may be released to the public. Date: December 22, 2017 Introduction The Census Bureau has provided estimates of the Citizen Voting Age Population by Race and Ethnicity (CVAP)1 and data to support redistricting under Public Law 94-171 (PL94) and Section 2 of the Voting Rights Act.2 This paper examines alternative sources for the citizenship data, specifically the addition of a question on the 2020 decennial instrument or the integration of administrative records on citizenship into the 2020 Census Edited File (CEF). In 2011, when they were released, the PL94 data from the 2010 Census had a reference date of April 1, 2010. The CVAP data released in February 2011 were based on the 5-year American Community Survey (ACS) data from 2005-2009. In addition, the 2011 CVAP data

1 https://www.census.gov/programs-surveys/decennial- census/about/voting-rights/cvap.html 2 https://www.census.gov/rdo/data/2010 census.html

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 143302

were based on Census 2000 block group geography while the PL94 data were based on 2010 Census block geography. The difficulty in integrating these two data tools for redistricting and enforcement of the Voting Rights Act was directly cited by the Department of Justice in its December 12, 2017 letter to Dr. Ron Jarmin, who was performing the non-exclusive functions and duties of the Director on that date. Data from Household Questionnaires and Administrative Sources The Census Bureau currently has four surveys containing citizenship questions. Citizenship is collected on the American Community Survey (ACS), the Current Population Survey (CPS), the American Housing Survey (AHS), and the Survey of Income and Program Participation (SIPP), and all persons in the household are in universe. The ACS, CPS, and AHS distinguish between citizens born in the United States, in U.S. territories, abroad to U.S. parents, and of foreign nativity but naturalized. SIPP collapses citizenship into a binary indicator of whether or not one is a citizen.3 Table 1 shows how much of the 2010 Census these sources cover. By linking citizenship data collected from the household surveys listed below to the 2010 Census, we can identify directly reported citizenship for approximately 14.4% of the total population. The integration of these surveys with the 2010 Census is based on the Protected Identification Key (PIK) added to all files using the Person Identification Validation System (PVS). From 2000 to 2015, a small number of the PIKs in these surveys do not match records in the 2010 Census. The nonmatch rate is less than 0.2% for all years, with a minimum of less than 0.1% in 2010.

3 This information is from the Master Demographic Pilot Report.

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 82 CU•APPENDIX 05 • ACLU • 78228

Table 1. Citizenship in Household Surveys Linked to the 2010 Decennial by Demographics

(DOCS 29-32) Household Surveys Linked to 2010 Decennial 2010 Decennial Noncitizen Citizen Missing Total N (%) N (%) N (%) (%) N (%) Total Population 1,523,000 43,090,000 1,192,000 100.0 308,745,538 100.0 Coverage 14.4 Sex Female 785,000 1.7 22,380,000 48.9 613,000 1.3 51.9 157,000,000 50.8 Male 738,090 1.6 20,710,000 45.2 579,000 1.3 48.1 151,800,000 49.2 Race 144

White 729,000 1.6 35,320,000 77.1 837,000 1.8 80.5 227,200,000 73.6303 Black 127,000 0.3 4,157,000 9.1 172,500 0.4 9.7 40,400,000 13.1 American Indian, Aleut Eskimo 14,800 0.0 562,000 1.2 15,780 0.0 1.3 4,007,000 1.3 Asian or Pacific Islander 364,000 0.8 1,688,000 3.7 93,000 0.2 4.7 16,770,000 5.4 Other 286,800 0.6 1,358,000 3.0 74,650 0.2 3.8 20,400,000 6.6 Ethnicity Hispanic/Spanish 675,000 1.5 4,046,000 8 8 198,100 0.4 10.7 50,480,000 16.4

Non-Hispanic/Spanish 848,000 1.9 39,040,000 85.2 994,300 2.2 89.3 258,300,000 83.7 Source: 2010 Decennial Census and Master Demographics, U.S. Census Bureau. Note: Household survey data unweighted. The reported population total is the official count from the 2010 Census. All other counts have been rounded. AL 3/23/19 145304

The Census Bureau has acquired multiple national administrative record sources that include citizenship data, as shown in Table 2. Table 2. National Administrative Record Sources with Citizenship Fields Currently In Census Inventory Universe Social Security Administration Quarterly Transactions Numident Temporary Assistance to Program Applicants Needy Families Bureau of Prisons Federal Prison Inmates Potential New Acquisitions Universe USCIS Citizen Data Population Real ID Act Data Driver’s License Applicants FHA Loan Applications Loan Applicants State Department Students studying aboard Expatriates and embassies registrations Medicare/Medicaid Program Applicants Loan Applications

Whether or not citizenship data are collected on the 2020 Census questionnaire, it would be consistent treatment to use administrative records to edit and/ or impute the citizenship variable, when necessary. From the sources in Table 2, the Census Numident is the most complete and reliable administrative record source of citizenship data currently available. The Numident file is a record of applications for Social Security cards. Unique, life-long SSNs are assigned to individuals based on these applications. A full record of all changes to the information (such as change of name) is also maintained. To obtain a

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 146305

Social Security Number, the applicant must provide documented identifying information to the Social Security Administration (SSA). Through the “enumeration at birth” program, children can be issued a Social Security Number (SSN) when they are born. Examples of data elements on a Numident record include name, date and place of birth, parents’ names, and date of death. As shown in Table 3, 90 percent of persons in the 2010 Census can be matched to the Protected Identification Key (PIK).4 Once a PIK is assigned, virtually every record is matched to the Census Numident (>99%). Nearly all the PIKs not in the Numident are Individual Taxpayer Identification Numbers (ITIN), which are held by noncitizens for IRS tax filing reasons. Among persons with non- blank citizenship in the Numident, 91 percent are U.S. citizens. Around 21 percent of the Numident records have a blank for citizenship. The Social Security Administration did not require evidence of citizenship until 1972.5 Many older persons thus did not report citizenship when applying for an SSN. We investigate this issue further below.

4 See NORC (2011) and Layne, Wagner and Rothhaas (2014) for details about the process used to assign and the quality of the PIKs used in data linkage at the Census Bureau. 5 A detailed history of the SSN is available at https:// www.ssa.gov/policy/docs/ssb/v69n2/v69n2p55.html (Exhibit 1).

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 82 CU•APPENDIX 05 • ACLU • 78228

Table 3. PIK Coverage of the 2010 Decennial Census and Numident Citizenship (DOCS 29-32) Distribution Count Percent of Percent of Decennial Matched Population Sample No PIK, not sent to PVS 10,367,975 3.4 No PIK, failed in PVS 19,198,234 6.2 PIK, but not in Numident, not ITIN 8,871 0.0 147 PIK, but not in Numident, is ITIN 1,566,645 0.5 306 Blank Citizenship 57,914,337 18.8 20.9 U.S. Citizen 200,422,211 64.9 72.2 Legal Alien, authorized to work 18,198,545 5.9 6.6 Legal Alien, not authorized to work 444,727 0.1 0.2 Other 259,674 <0.1 <0.1 Alien Student, restricted work authorized 184,673 <0.1 <0.1 Conditionally Legalized Alien 179,646 <0.1 <0.1 Total 308,745,538 100.00 100.00 PVS is the Person Identification Validation System. AL 3/23/19 148307

One of the reasons why some person records fail to receive a PIK is insufficient personally identifiable information, which is the case for the 3.4 percent of records not sent to the Person Identification Validation System (PVS), as shown in Table 3. It is thus likely that many of the same records for which it is not possible to link in citizenship information due to a lack of a PIK also have imputed values for other demographic variables. Tables 4A-4C show that imputation rates are much higher for 2010 Decennial Census person records lacking a PIK, especially for date of birth (a characteristic which may be hard for proxy respondents to report on behalf of their neighbors, for example).

Table 4A. 2010 Decennial Census Gender Source, PIK vs. non-PIK Records With PIK No PIK As reported 98.7 75.4 From first name 1.3 1.4 Value edited for household consistency <0.1 0.4 Allocated from hot deck 0.0 2.1 Allocated from consistency check <0.1 <0.1 Substituted <0.1 20.7 Percent of Sample 90.9 9.1 The number of observations is 304,450,000. Group Quarters are excluded.

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 149308

Table 4B. 2010 Decennial Census Date of Birth Source, PIK vs. non-PIK Records With PIK No PIK Fully reported date of birth 96.0 35.7 Only day of month allocated 0.2 0.5 Month and day both allocated 0.3 1.7 Year of birth created from 0.7 0.5 two-digit year DOB allocated consistent with 1.6 16.4 reported age DOB allocated consistent with 1.3 24.7 allocated age Substituted 0.0 20.7 Year of birth of householder or <0.1 <0.1 spouse adjusted to be consistent with number of children Percent of Sample 90.9 9.1 The number of observations is 304,450,000. Group Quarters are excluded. Table 4C. 2010 Decennial Census Race Source, PIK vs. non-PIK Records With PIK No PIK As reported 96.6 68.6 Code changed through consistency edit <0.1 <0.1 Assigned race from response in Hispanic question <0.1 <0.1 Allocated from within household 1.5 4.1 Allocated from hot deck 0.7 5.9 Substituted 0.0 20.7 Assigned race from previous census response 1.2 0.7 Percent of Sample 90.9 9.1 The number of observations is 304,450,000. Group Quarters are excluded.

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 150309

The Estimated Effects of Including a Citizen- ship Question on the 2020 Census We also study how including the citizenship question might affect response rates by comparing first mailing response rates in the 2010 Decennial and the 2010 ACS for the same housing units. An important difference between the two questionnaires is that the ACS questionnaire contains citizenship questions, and the Decennial Census does not. Households with noncitizens could be particularly sensitive to the inclusion of citizenship questions. Here we focus on housing units that received a mailing (housing units in the initial mailing and that did not have mail returned as Undeliverable as Addressed (UAA)) and which were not classified as a vacant or delete. The housing units are divided into two groups, those where at least one person is a noncitizen in the Census Numident and has been assigned to this housing unit in the 2010 Census Match Study’s administrative records person-place (PIK-MAFID) crosswalk, and those where all of the persons are citizens in the Census Numident. Table 5 shows the 2010 Census and ACS response rates for these two groups. The self-response rate is higher for 2010 Census than for the ACS for both citizenship categories, presumably reflecting the higher burden of the ACS. The citizens6 response rate is greater than the noncitizen rate in each survey, suggesting that noncitizens have a lower participation rate in general. Most important for this study is understanding how the difference in self-response rate across groups varies between the 2010 Census

6 Citizens include those born in the U.S., those born abroad to U.S. parents, and naturalized.

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 151310

and ACS. While the self-response rate for citizen households is 13.8 percentage points lower in the ACS than in the 2010 Census, the self-response rate for households with at least one noncitizen is 18.9 percentage points lower for the ACS than the self-response rate to the 2010 Census, which is a 5.1 percentage point difference between the two categories. Though there could be other reasons why households with noncitizens are particularly unwilling to respond to the ACS, this evidence is consistent with citizenship questions being more sensitive for households with noncitizens. Table 5. Comparison of 2010 ACS and 2010 Decennial Census Response Rates, by 2010 Numident Citizenship Status Response rate Difference Row (%) Percent (Numident Status) Census ACS Citizen 79.9 66.1 13.8 94.1 Not Citizen 71.5 52.6 18.9 5.9 The sample size is 929,000 households. Other proxy measures for understanding response sensitivity to questions of citizenship can be examined with longitudinal data. Using the 2014 SIPP longitudinal panel waves 1 and 2, Table 6 shows household response rates for citizens and noncitizens. Noncitizens made up around 6% of the 2014 SIPP survey. Of persons living in households where at least one individual did not respond to the survey questionnaire, noncitizens made up around 8%.

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 152311

Table 6. Noncitizens and Non-Response in the 2014 Survey of Income and Program Participation Wave 1 Wave 2 (%) (se) (%) (se) Noncitizens 6.1 (0.144) 5.7 (0.096) At least one member in the noncitizen household did not respond 7.9 (0.473) 8.5 (0.351) Source: 2014 SIPP, Waves 1 and 2 Note: Citizenship status refers to status in Wave 1. To get a sense of the quality of the survey and administrative citizenship data, we compared ACS and Census Numident responses for the same PIKs. Table 7A shows that over 99 percent of the blanks are U.S. citizens in the ACS, so it is highly likely that persons with blanks for citizenship in the Numident are U.S. citizens. Among those who are legal resident noncitizens in the Numident, roughly 40 percent say they are U.S. citizens, nearly all via naturalization. This suggests that either the Numident citizenship data are out of date, or that there is a tendency for noncitizen ACS respondents to report being U.S. citizens. To provide context for these discrepancies, note that the share of the stock of legal permanent residents who became naturalized citizens was 8.3 percent, 6.0 percent, and 4.9 percent in 2008, 2009, and 2010, respectively, suggesting that the Numident data would need to be several years out of date to explain the observed discrepancies, if the ACS data are accurate.7 If the Census Bureau obtains the U.S.

7 The data on naturalizations come from https://www. dhs.gov/sites/default/files/publications/Naturalizations_2010.pdf,

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 153312

Citizen and Immigration Services (USCIS) citizenship file, we would be able to measure how up to date the Numident citizenship information is. One way discrepancies can occur between the ACS and Numident citizenship information is incorrect PIK linkages. In Table 7B we include only PIKs that have median or above PVS scores in the linking attempt matching on the most information (geosearch pass 1). The discrepancies are smaller for the cases where the PIK is a citizen in the Numident, but they are larger where the PIK is a noncitizen in the Numident. This suggests that the significant discrepancies with the ACS when the PIK is a noncitizen in the Numident are not due to linkage errors with the PIKs.

and estimates for the stock of legal permanent residents come from https://www.dhs.gov/immigration-statistics/popu lation-estimates/LPR.

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 82 CU•APPENDIX 05 • ACLU • 78228

Table 7A. Comparison of 2010 ACS and 2010 Numident Citizenship, All PIKs ACS\Numident Blank A=US. B=Legal C=Legal D=other E=Alien F= Row (DOCS 29-32) Citizen Alien, Alien, not Student, Conditi- Percent authorized authorized restricted onally to work to work work legalized authorized alien Yes, Born Citizen 95.1 96.6 3.8 3.6 11.6 2.0 5.6 91.0 Yes, Naturalized 4.1 3.2 36.9 37.3 11.0 47.2 42.4 5.3 Not a Citizen 0.8 0.3 59.3 59.1 77.3 50.7 51.9 3.7 Column Percent 24.8 69.5 5.3 0.1 <0.1 <0.1 <0.1 The number of observations is 4,022,000. 154 313 Table 7B. Comparison of 2010 ACS and 2010 Numident Citizenship, Higher Quality PIKs

ACS\Numident Blank A=US. B=Legal C=Legal D=other E=Alien F= Row Citizen Alien, Alien, not Student, Conditi- Percent authorized authorized restricted onally to work to work work legalized authorized alien Yes, Born Citizen 97.2 97.9 2.7 2.8 26.4 2.4 8.6 95.4 Yes, Naturalized 2.4 2.0 44.0 40.9 17.4 45.8 47.1 3.1 Not a Citizen 0.4 <0.1 53.3 56.4 56.2 51.9 44.3 1.5 Column Percent 28.1 69.5 2.3 <0.1 <0.1 <0.1 <0.1 The number of observations is 2,168,000. Only PIKs with a median or above score in the Person Identification Validation System (PVS) geosearch module pass 1 are included here. AL 3/23/19 155314

Table 8 shows the ACS citizenship response distribution for ITINs. About 7 percent report being citizens, though only noncitizens should have ITINs. Table 8. 2010 ACS Citizenship Responses for ITINs ITIN Yes, Born Citizen 4.9 Yes, naturalized 2.4 Not a citizen 92.7 The number of observations is 42,000. We next examine how the discrepancies between the ACS and Numident citizenship responses vary by whether the household responds to the first mailing vs. different kinds of follow-up. We restrict the ACS sample to the population of individuals in households that received a mail-in form A self-response in our sample refers to an individual being part of a household that successfully responded to a first ACS mailing. An individual is classified as a “Mail Follow- up” (Mail FU) if that person responded to a follow-up mailing. Lastly, an individual is classified as CATI/ CAPI if that person did not respond to the initial mailing and ended up receiving a telephone or in- person follow-up interview. To assess the reported citizenship in the ACS, we consider individuals in our ACS sample who also match to the Numident, giving us an additional source of citizenship information In the Numident, we classify all citizen categories as well as missing citizenship as citizens, for the reasons given above. Table 9 shows the distribution of ACS outcomes for individuals who are also classified as citizen or noncitizen in the Numident. Regardless of the response mode, individuals classified as citizens in

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 156315

the Numident also reply that they are citizens in the ACS, while nearly half of those classified as noncitizens in the Numident report being citizens. Thus, the patterns shown in Table 7 vary little by response mode. Table 9. Comparison of 2010 ACS and 2010 Numident Citizenship by Response Type ACS\Numident Citizen Not Row Citizen Percent Citizen, (Resp.) 63.9 22.7 61.6 Not Citizen, (Resp.) 0.2 29.2 1.8 Citizen, (Mail FU) 21.1 9.9 20.5 Not Citizen, (Mail FU) 0.1 13.9 0.9 Citizen, (CATI/CAPI) 14.6 8.5 14.2 Not Citizen, (CATI/CAPI) 0.1 15.8 1.0 Column Percent 94.4% 5.6% The number of observations is 3,752,000 individuals. Mail FU is Mail Follow-up, CATI is Computer-Assisted Telephone Interview, and CAPI is Computer-Assisted In-Person Interview. Other Potential Administrative Record Sources of Citizenship Data There are several additional administrative sources of citizenship information that the Census Bureau could consider trying to obtain. Most important are the USCIS citizenship and noncitizen legal resident files. The citizenship file could be used to evaluate the quality of the Numident citizenship data, and in particular, how quickly it is updated. The legal non- resident file could serve as an additional reference file, so that more noncitizens can be given Census PIKs.

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 157316

Another likely useful source of citizenship is state drivers’ license data. The REAL ID Act of 2005 requires evidence of citizenship to obtain a driver’s license. This has been fully implemented in 28 states, and the others have waivers. The Department of Homeland Security is overseeing implementation of the law. Starting January 22, 2018, passengers with a driver’s license issued by a state that is still not compliant with the REAL ID Act (and has not been granted an extension) will need to show an alternative form of acceptable identification for domestic air travel to board their flight. Each state must agree to share its motor vehicle database with all other states. This database must include, at a minimum, all the data printed on the state driver’s licenses and ID cards, plus drivers’ histories.8 These databases could become an important source of citizenship data. Other potential sources include FHA loan applications and Medicare and Medicaid applications. It would also be useful to obtain data on U.S. expatriates from the U.S. State Department. The State Department may have data on students studying abroad and expatriates registering with embassies. These data would prevent these PIKs from being mistakenly included in the administrative record person-place crosswalk. It is worth noting that others who are interested in noncitizens have used administrative records to estimate stay rates and other relevant characteristics.

8 There is some debate about whether a national database is being created from these data. DRS says this isn’t the case, but see https://papersplease.org/wp/2016/02/1/how-the- real-id-act-is-creating-a-national-id-database/.

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 158317

Finn (2014) developed stay rates for students in science education by linking social security numbers of students enrolled in science programs with the Internal Revenue Service (IRS) tax records. Not only is using administrative records potentially a more accurate measure of citizenship, but it is also cost efficient. The Bureau already acquires SSA Numident information on a quarterly basis. To collect that information through self-report by adding questions to the 2020 decennial would require additional unnecessary costs and burden to the Bureau. Implementation for the 2020 Census The direct solution to supporting redistricting in the manner requested by the Department of Justice is to make a citizenship variable available on the 2020 Census Edited File (CEF), the internal, confidential data file from which the PL94 tabulations are produced. If citizenship were available on that file, the PL94 tabulations could be restructured to include direct estimates of the citizen voting age population by race and ethnicity at the block level. These tabulations would have essentially the same accuracy as current PL94 and Summary File 1 (SFI) data. We recommend provisioning the citizenship variable onto the CEF by record linkage using the national administrative data discussed above. Once the citizen tabulation variable is added to the CEF, it would be available to the 2020 Disclosure Avoidance Subsystem (DAS) for inclusion in a modified version of the proposed P2 table “Race/Ethnicity for the Population Age 18 and Over” where “Population Age 18 and Over” would be replaced by “Citizen Population Age 18 and Over.” This revision would allow the use of the same

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 159318

disclosure avoidance methodology, state-of-the-art differential privacy, currently available for the 2018 End-to-End Test and the enhanced methods, integrated PL94 and SFI protection, planned for the 2020 Census itself. This version of P2 would be the first PL94 data produced at the block-level with estimates of the citizen voting age population by race and ethnicity and with accuracy comparable to the accuracy of the P1 “Total population” table. The P1 and P2 tables would tabulate race and ethnicity in the same manner as currently proposed. Tables P42 “Group Quarters Population by Group Quarters Type” and H1 “Occupancy Status” would not be modified. The 2020 Census questionnaire would not be altered, and the field operations would not have to be expanded to compensate for the lower rate of voluntary compliance predicted for a census that asks the citizenship question directly. References Finn, Michael G., 2014, “Stay Rates of Foreign Doctorate Recipients from U.S. Universities, 2011,” Oak Ridge Institute for Science and Education. https://pdfs.semanticscholar.org/7a4c/49e7878730bS8 7201548338aa6052e2401b7.pdf. Layne, Mary, Deborah Wagner, and Cynthia Rothhaas, 2014, “Estimating Record Linkage False Match Rate for the Person Identification Validation System,” CARRA Working Paper #2014-02. NORC, 2011, “Final Report: Assessment of the U.S. Census Bureau’s Person Identification Validation System,” University of Chicago: Bethesda, MD.

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 160319

Summary of the process for using administrative records to add citizenship to the 2020 Census microdata and produce block-level tabulations from the enhanced data

Background: Currently, the Census Bureau produces the PL94-171 redistricting (PL94) data, which are block-level tabulations of total population and voting-age population by race and ethnicity using the OMB-approved categories, from the decennial census. These data are mandated by 13 U.S.C. Section 141, and must be released by April 1st of the year following a decennial census. The Bureau also produces the Citizen Voting-Age Population by Race and Ethnicity (CVAP) data, which are block-group-level tabulations of the citizen voting-age population by the same race and ethnicity categories. These tabulations are produced from the most recent five-year ACS data and released annually in February. State redistricting offices use PL94, CVAP, and other sources to draw new Congressional and legislative districts that conform to the one-person, one-vote mandate and the Voting Rights Act. The Department of Justice uses PL94 and CVAP to enforce the Voting Rights Act. Both products were designed with extensive input from their stakeholders. Because the PL94 data are tabulated from the universe person-level Census microdata file, they are treated as exact block-level tabulations, although they have been subjected to the Bureau’s disclosure avoidance algorithms, known as confidentiality edits. The CVAP data are based on the ACS sample, and are released with block-group-level margins of error. The statistical fusion of the PL94 and CVAP data

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 161320

to produce block-level citizen voting-age population by race and ethnicity necessarily involves modeling, and produces block-level estimates that may have very large margins of error inherited from the CVAP block-group-level MOEs. Request: We received a Department of Justice request to produce more accurate block-level CVAP tabulations by asking the citizenship question on the 2020 Census, and incorporating those responses into block-level tabulations instead of using the ACS data. We believe higher quality dataproposed linking an administrative record-sourced citizenship status variable directly to the 2020 final edited microdata file, and tabulating CVAP from those data instead of putting citizenship directly on the 2020 Census questionnaire. Method: We currently use the PVS system to add a PIK to the confidential 2020 Census universe person and household microdata outside the decennial census production system. The current decennial census production system accepts these PIKed data for downstream processing. External to the decennial census processing system, we would create a universe “best citizenship” variable from NUMIDENT, United States Citizen and Immigration Services (USCIS), and possibly other sources, using methods similar to the ones we have used to link other administrative variables into production tabulation systems. The “best citizenship” variable would be linked to the confidential 2020 Census person-level microdata record using the PIK. Tabulations and Disclosure Avoidance: A version of the redistricting data would be created tabulating citizen voting age population by race

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 162321

and ethnicity at the block level. These tabulations would be protected using the new differential privacy disclosure limitation system being implemented for all 2020 Census tabulations. The differential privacy system provides global confidentiality protection guarantees that account for the information in all 2020 publications and are provably resistant to all future external data at a level controlled by the global confidentiality budget. The new block-level CVAP tabulation would share the global confidentiality budget of the 2020 Census, not augment it. Whether we would release this product as a substitute for voting-age tabulations or a supplement has not been determined.

Abowd to Jarmin January 19, 2018 Pre-decisional

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 163322

Question Assigned to Completed 1 Reist 2 Reist 3a Abowd: Mule X 3b Velkoff 4 Velkoff 5 Abowd: Mule X 6 Abowd: Brown 7 Abowd: Brown X 8 Velkoff 9 Abowd: Brown 10 Abowd X 11 Abowd: Brown 12 Abowd: Berning X 13 Abowd X 14 Abowd: Berning X 15 Abowd X 16 Abowd: Berning X 17 Abowd: Berning X 18 Abowd X 19 Abowd X 20 Abowd 21 Abowd X 22 Abowd: Berning X 23 Abowd: Mule X

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 164323

24 Abowd: Brown X 25 Velkoff X 26 Reist X 27 Abowd: Brown 28 Velkoff 29 Velkoff 30 Velkoff 31 Velkoff 32 Reist: Whitehorne 33 Reist 34 Abowd X 35 Resit: Dinwiddie

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 165324 of be ily r his may t in s w necessa content vie r not he thei T do . . at h t and , Bureau hope community the author(s) h it Census w . the S . Mechanisms of U statistical made the is Access those of e r general a Data the those Methodology to esentation r Conference and p is Census esentation Experiences r epresent Private p interest r Th Bureau 15am : Policy Garfinkel Statistical . L on and Census and 30-10 . : Confidentiality 2020 8 S , . Simson U Research Committee the Scientist 018 2 to Federal Senior Differentially Challenges Commerce of ~f,1h,1 Department ,1 c U.S. ,ensus.QO,' U.S.CFN

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 166325 used the shall to or records. agency of or and nor the and or the any of public of title, individual. for 210 in, individual than or this evidence bureau or employee of exception process, ager19'.thareof, 5 or 1998. as or censuses other section or require, provided by or legal Act, particular individuals chapter title shall bureau officer, for of from admitted purpose or establishment this covered Ill establishment title, of data Department customarily be any proceeding. of or 10 such the this agency, Appropriations for immune subjects of of Commerce from, any subchapter be the title of chapter particular by by to confidential; or concerned, bureau, Agencies this as for shall 16 any compiled of or purposes administrative as title by is 8 employees retained or Deparbnent Related the this confidential treatment that provided of and the retained, and the 5 out section of been department, to in provisions so establishment judicial or furnished No therefore privacy! or the officers have chapter Judiciary, been data relating other governments carrying of employee provided of in IV the or the or under obtained supplied; as which reports. sworn have or is section individual Information suit, State, it the officer this except the censuses protect which and reports of subchapter whereby Secretary 9 - of than other the information furnished individual which action, may, (a) by identified; to to the any any for for the Justice, be any other reports, to census nor liaison apply in consent of can publication except not subsection Commerce provided the f information respect o of need title anyone census examine census Code§ purposes any shall Commerce, copies the Secretary, data to of purpose with of this the Department we without Use Permit Make any U.S. census.!)()',' ~~~~~•:sAdnMr>,tra!KJn provisions current I U.S. · (1) (3) reason, thereof Government, (2) under not, statistical Copies for government Neither The e,areau States But, 13 Ca) (b) rocal interim Departments governments, establishments, United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 167326 Census 2010 · the HANDS for OUR IN 2010 IT'S census Avoidance Commerce of Department . U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · Disclosure e,areau States United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 168327 by are quick easy, your is law." protected and and answers "It es. MII 2. - □ trac and 7 ., oo. ~rson ..,..,.Tcnp,, .,.,.,_lr-.-_ oo origin PM J.'d 1olbirth? Yl'llh 7 - ...... dat Pacificlslaod9< i,, IUspanlc Yearolbinh origin? C=::J o,.,.._,_ ..... na~homlll FoHawaiian lspanlcoriginsareno 1, □ 0 O •boll! 0 H □ □ □ 0 8 Spanish Day oooormM1b:u'es. D ONEOO.. >' .or 1tlsi>«GOn1'a 7 e<~on OutttioP ~ no ln Ou H Pmll)OO?bei5.i'lt:io:.w. Lati l.1uil .,..,, ~ . J__., Kor&M Llbno,orSjlanoohorigin v- ce?Ma/ll. , ... ilOi 91andwf) a u?Marl<, · ,;oun~ □ □ □ 0 Ptitl1K1,lb< Ma.ucanAm.,Chocano 's, 'ss 's1 1 15panic , 1 Aia!IMSOMI lncolegehouslng ln!hemiilary o,..,.,,,.i,- - N 20 were Female anum . YM-Mllrl<.tal,;,,,tlJ()ply. □ D O ~------~ ~ O lol , e™>n □ ~™>n1 Mc.!Hillpanic people ,. erson1sometlmesllveorst11yso"'"'hereel11e? , , P , is P so~ is ls No YM Vn,Clban AmericanlndianorAlaskaNalM-l'Yir,_,,,ot..,.,kd,rixq,,Jlflt7 ~ Bilck."'fric.-iAm.,orNegro O!hw...... , Someolherra.ce-Prinfraoe.7 ..._..,lm,g,W..,Thai; ,.._ - a.na.a No YM Yn Mianlndian Male A~Cotnillo:,,~~~~. Pe< 0 ; more What le /tOTE: What What H ~""~1 P!ea6'!repMDabie&asagoo~rh9ctiidillGsslllall!ye,arold. OU61ion9a!lou1race.Forlhlscensus, □ □ □ □ □ 0 O O 0 0 0 c::::J ,Q § .Does 9. ➔ 6. 7. 8. ➔ 10 ' 1 o a a heor ngin , i . tw,ce e 2010? CWlted OfierwisQ ~prisoo . . 1, be orli\l,._.,bal>ysitle< .. lhllnuminghoma,lha coumalnlnstltutlon may April , lncludelnOUeSl!cn1? s l c=i lhey epeoplewithoutapermanenl , . lud --- ,iomeon □ . ,onApril1,2010. . rd.dngblll»es,who!iveandsl99phwe Othe!wlse 2010,COIHlha1p,non ~not : 2010, , c:::::J here , ouchas-.nt>Abitiorf/,ti"'-.., Ille. , ormoblle , M1hoi.lamongi,gaorl0An)? ,so: ( ial, .,, smustel,oinc eorbltickpen u ~r:7::-:::.::.mber"!W<1111i1ycalifw9 1 Start elhefeanyaddhlonalpeoplestaylngho,, Nonrelatr;tt,MlChU"'°"""l'J Peop1a~ng""'81emp,,,arjy ""°additionalpeop1e Relatives,suct,asadl.ftcnii,;!,e,',,ocusi.,s,orin-laws Chlcnn Ow,>idbyyouor-~lfllhrlhouMholdl,...ond nolau,tanyonalivingawayeilherato:,lege«inlha mortga,geotloan?lncludehomeequi1ybans dear Occupedwitlidbyyouor-~inthishouNlddMlli .... r prll1,20101h AreaCodll+Nlnber malar), Mlh&nlaf!,..th9yleave""""9e shemaybGrnis.s.edinllecensus. deteotion!acity,elc Marl<1drlvlt4'P/y her8onApril1 - Numl>erot,-,,,i.~ A □ □ HowmanypeoplewNelMngorstaylnglnthlsl'H)use, apartment Oonolcot.rllanyoneinanur$S)Qhome O ':ia~•,"°o".:-·:::.rtment,ormobilehome- ltsornoonewhotwnopo,maf'IOlllp1¥&tostayi&S1a}'Wlg LOOV81118Sepoopkloll)'Oll'form,QWl'lifth9ywillrowmkl c::::J :::; □ ~~a1.:.::::: 0 O O O o lsho

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 169328 11 7,987,323 116,716,292 308,745,538 300,758,215 Results Basic population Population: of population quarters Commerce of population Census Department . U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn Household Households Group Total I · e,areau States 2010 United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 170329 2 17 115 126 600 values 73,768 492,660 12 10,620,683 Distinct schema Categories) database (0MB 1 Categories) cells Categories) person (SF2 (0MB High-level to blocks tracts histogram Commerce of Census: Department . National Race/Ethnicity Relationship Habitable Race/Ethnicity Habitable Sex Age Variables U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · e,areau States 2010 United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 171330 25 zeros) counts 13 approximate) 30,874,554 2,771,998,263 2,806,899,669 2,093,683,376 7,703,455,862 are Released (including (counts statistics 1 File publications published sample 2 on the Summary micro File Redistricting of of Commerce bound of Census: Department . U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I Balance Lower Publication PL94-171 Public-use Summary Statistics/person · e,areau States 2010 Summary United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 172331 . n j . "r a_v ltt ,,,. !):lo ,·I ,1,,. II•• ~•i«!, r,L,·· 'I""".,._.. p,,nurb..-- ,,..., ~,i1>ing ,.,a,1...,._.,,r,. I d :"J. ,u "'""'· 1_,,.....,L_. . .. "'"""lurr (iiil , ,..,..ly .,,,,-~ l rdatmel,·•na.>Jl rn" f""'o tb.• c-.o•ei:....-ie.. 1 1.,-j\;orV • .ri1,g aod •"h .,kn.,..,. l Privacy ■ <.f W,.,.,,d,m "l""".J • n•u d• u1,11rn ;, ,;f,t,,-;aj,.,,.., ..,. d,.,.;a;kl Ja111.,;,,.e • ,...,,...,y ,u J~-U""'~' 1 ,1u,.,.,.,n.,_,..., ..,._1 a otheque,-,·r-rictio:napproach l ' .. ,.,,...,,, fl ,.....,. ,.,.,. .. · • alJo.r.b- . r; 1b,c.- , ~ ~t:'~~~~ ·u•,1 ■ m,:(~in.o;,hequ,ryaa:udm~ly. """" lt ~11,..,..i,""' - •Pl>t""i•n.,t<"~g,.".~JJl"'"'";m.o,.- l to~ ,;l 1 t,.-,. ·2 ,-p,:,..ifi~ • ,\darn ,,r penwbad1"<1:-.l~,,.P<..,. •U •trinioo. •nur· --1,,,, ...,.i -,- 'V'""" NJ08540 , Kobb while . , , • lo r.. 1 fot , '" S.b-- ..__. @research 1 l J ,. h~,od. i nur ,.,,...,.,._ i Prmceton __ 1 4lndependeoceWay ~,ai,.,~,. ,.~ NECResearcn D .. r,bu«l .... u,J,,,.. 11.,, , •~• ,...,,.,,..., 1 rit 11 I • 11....-h;,,oi,-u, l u( DIM.V' 1,.. , • ti,~ ,r, .11 TOf.o1100.-,<>f111t1- 1 .,.,,,,_.,.....__,,_,,,«6: to'-,, ..,.,1.,,•~••,L, __ 11• l•..fi"I0-6-0•06 .. .. 1 pro1< l l•"'1>1t.-,J ••• ••lu~ 00J ;,, a·I""' .. •i a1i.-..,...-,.,;,,.,. - IWl-oo•I>< 1,.,.,1 Thr •••I .. 1;-.; r.:, ..,.. .._,... .. ot.li,.- ~ .• , ~ · - .,_ ..,.,1,m1m~...,. • ... Revealing • -~ ... .,.,,,_r .. •• .-1,.-•., • ·•dl 1••l•·U1 ,11,,nnal•M1 ,.,.,.t, I"''~•• "'1 TRJ\C."T l• ,~,:;;;;.;;.~· ,1;,,,.1..,.,1,, k-ali ·1f~· ...,.1, . the BS .. 1, A •<-. ,.,. •Ii• • "W,«l ...,.,,_ _..,._,,- ~ '"1~• f'OD>.;oo_,_,-,_ Cop)npl!OIHKMl ,q:,ubl,>L0>p,,,aw""'"' ,,f ,.,.,..,....,.,..1., "''OJ:~' ,...... ,...o,dw,i,-.,...,,.- ...... -- of to an re- . sta sub That [n] by to privacy privacy vn) <;;; preserve violation a q exemplify magnitude 0( usability that by O(vn). polynomial database achieve strong and a (perturbed) subset algorithm a to is a tight in is magnitude demonstrate noisy databases perturbation privacy order result of being we statistical magnitude in a results from result of main query whose that this a reconstruction between data model statistical Our always of adversaries for this We show perturbation that with d;. Reconstruction Commerce we algorithm dn, tradeoff of perturbation , E,Eq show .. , adding bounded the algorithm by d1 add We Applying access perturbation algorithms databases. . Department to while databases U.S. [l'fl'il6.gOY ~~~~.,"""-w•~ time--T I string ,/T. · has examine access sums. answered smaller privacy For ates t eurem S ing is, construction be privacy preserving n-bit ABSTRACT We tistical of one is~ statistical set Database 2003: ted i n U ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 173332 i ' .. !! properj a funct addit Datal individ, applying Smith the """' il of 1eam holds oft.c-. privacy the anplying query mapping c. ■ by a a ....., of / oo of Fello,nhip. that Adam result 11] laa - Private input ~ the • 10, and remlt [ i,erwr , privacy lr, = function On 2 perturbed in in ;'l the • • is .bgu.ac.il the ■ i ■ Postdoctoral query trusted ...... -ii. Nissim an.tweris ~~ server. = . a 11, initiated a' answer w im&n -i ~= O.n.ttllo!i' tnie er kobbitc ~.aaithlveizllanll.. a {dvork,•cllll.erry}l:alicro P .. Kobbi protecting true GiWII trusted , ~ 1 M. a reaearch Consider the a Sensitivity of by Valley. Science. while so-called l!cal!mlllii'~ Analysis Anit University. of line to held the a 11D-called McSherry is privacy, and whole Silicon in!ormation...... datab-. . ,al!!cmll'!laa a the L re!WI, CL l!!e.11 a Institute as l!!c Frank to continue Noise , Ben-Gurion protect reals. 1 Louis datab~ semri.tive ~ Rellearch, We statistical. to To ~ o! the the Dwork by Wei.m:wm . wwwll!l\llemgl!Pla i, databaaee population s l!lp,!l , ' ,, Microaofl. assume the Abstract. E; , ~ to< databMe datahMM dat, database. preserving 1 of Cy:nthia We the Supported privacy- mapping ataba11e Calibrating * / ties a oontributor l Ito a = of /, be by its re / devi maps to chosen holds g privacy terms Roughly returned applying mapping change /. is / in on of which mechanisms applications functions and that perturbed separation can 11] standard in is / carefully understood noise, a result the to [10, server privacy function function general obtain to in sums, of database the plus particular answer to the we sanitization is of the query trusted noisy previously true argument a a of calibrating initiated answer several study of according the answer was by for row the single true interactive Given case sensitivity Additionally, true of than ith research characterization that Consider any the the privacy, the of generated extend the preserved that to value on clean line Privacy shows needed be We a noise so-called is protect response, information. transcripts. very databases. l]. can denotes the amount a of Commerce To [O, focused increased of according this noise analysis is x, the random continue to is the reals, less privacy and of new work sensitive step noise We statistical to where rows this of Department database. . the that The first user. U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn of showing the I non-interactive. addition case. · g(x,), Previous The the , to e,areau f proving preserving Abstract. indistinguishability database databases distribution, database output. over ation speaking, substantially sults L the to the States Differential 2006: United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 174333 II , . P ■ ■ ll ■■■ • ■ fSRDC 16 File: System Edited Avoidance Census the on Disclosure filter, 2010 a and as Commerce of Department . 2000 U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · e,areau States operated The United ensus ~•ttttif! t-t~titftt!t itl,~\ttl --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 175334 on 17 "X" relied State 2010 and state data secret. a 2000 in within processing remain external are of must used lack Census swaps of against the if on rest swapping attacks privacy guarantees to of protect system counts based or formal state details swapping: households: "invisible" of state-by-state Commerce and is of reconstruction consider provide affect understand guarantee run rate not not not to Department be . protection U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · Privacy Does Does Easy Does database Swap Operation Can e,areau States • • • • • • • • Disadvantages: The swapping Advantages United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 176335 privacy. formal 2020 census embraces I. . R , y End - nt Bureau u to Test o - C c : End ldcm Commerce v of Census Census Pro 2018 Department . US U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · e,areau States The United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 177336 not 19 serious from a were is responses systems Census 2003) survey attack. published 2010 of Nissim the tabulation to kind and individual summaries this of (Dinur statistical applied acknowledge. tabular all against edits that must privacy person per defend the data threat independent reconstruction Commerce to of Census: Department . billion confidentiality statistics U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn protect I · 7.7 25 e,areau States Database • • Motivation: The To 2010 designed disclosure confidential United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 178337 20 File System" Detail (2020) Microdata system. tabulate." Avoidance to "safe records production are System that Disclosure Avoidance "Disclosure a Census Census records the edited DAS: create File the make to the into to on is Commerce of Census of Edited Department . drops plan U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · Designed Operates e,areau States • • Features Our that United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 179338 II , . P ■ ■ ll ■■■ • ■ fSRDC 21 Census the protections. allows , Budget Decisions System Privacy confidentiality Accuracy global Avoidance enforce Commerce of to Department . Disclosure U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · e,areau States Bureau The United ensus --- ( n,tt.ntt! t-MtltttOt jtl't~ittl

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 180339 '"' E) ( ,;, (DSEP) budget 22 ,:,, loss ROCcurve differential Committee .,;, Privacy use (ORB) Policy attack, om will Board Executive ~ u u (ll (ll :::; ()' ro 0 Review system Stewardship Disclosure reconstruction Data from a by released accuracy avoidance privacy. trade-off policy the the provides: database and determined against the on recommendation allow given budget with disclosure privacy that possible decide defend bounds accuracy Commerce of to to best privacy-loss Department . Census U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn the Final I · reconstruction makers between Provable of tabulations. Algorithms e,areau States • • Differential privacy The United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 181340 System Process £ 23 Global budget Disclosure Confidentiality Protection Avoidance infusing loss on accuracy privacy data best privacy: relies the for attacks in external formal once on at tallied System with be guarantees depend reconstruction must processed composable not infusion privacy data be do noise. are Avoidance database noise must tunable private of of Commerce private and against of understand guarantees operations country use to Department . Disclosure U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · Easy Provable Entire Privacy Every Privacy Protects e,areau States Disadvantages: • • • • • • • The formally Advantages United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 182341 SF- and 24 age, PL-94 produce MDF? non-voting to age, geography private stakeholders systems of (voting answers levels external totals all tabulation query at and differentially a legacy exact among population internal with Commerce to of generate Department . U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I tabulations · 1 householder) Familiar Operates Guarantees Consistency e,areau States • • • • Why United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 183342 data. 25 areas. swapping. raw MDF at by small accounted. on looking private and by advance. introduced in reports understood. done error tracked or data be processes: known the be to be known differentially know cannot accurate a need must business widely didn't tables data highly MDF not on is on Census 2010 want creating to challenges: and in checks privacy users queries confidential Commerce of 2000 of required in data quality Department . uses desired U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · Differential Many Data Users All All e,areau States • • • • • • Communications Changes Challenges United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 184343 queries 26 over distribution 18 distribution and distribution SF-1 18 under PL-94 and relationship ownership exact: ethnicity the meets protected" be and PL-94 distribution distribution not and Race Household Household Age Age 1 • • • • • Other "Privacy will privacy (but test" Section types vacancies E2E 2, & counts) population queries Differential 2018 quarters age for Commerce of population Article group exact: population householders voting be Department their . U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · Block Block Block not Certain e,areau States • • • • Specific PL-94 must "Invariants United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 185344 I. R . )', End - to Test ounl ~ C .~ End Census Provldcnu 2018 R.I. Test County, Commerce of End-to-End Department . U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I · e,areau States Providence 2018 United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 186345 100 DASO (E) o.7s 40 budget rve u c o!,, loss OC A Privacy 02' 000 100- 000- O.!il· 015- ~ u (.) (ll (ll :::; ()' ro 0 creating or are budget Questions? privacy Avoidance loss We accuracy no no engine engine Mode privacy Disclosure Test on privacy, accuracy, & avoidance) Commerce for yet System: Mode of 100% 100% "bottom-up" "top-down" requirements. - - - - Systems: Department . disclosure U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn I decision · (No DAS2 Development Production DAS1 DAS3 DASO framework e,areau States • • • • • • No accuracy Systems: Operational Testing A United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 187346 privacy 41 or budget differential produced be will Test privacy-loss incorporate files the will algorithm Questions? test End-to-End PL94-171 regarding Bottom-up 2018 yet - the End-to-End level Commerce prototype of DAS2 for the Department . 2018 U.S census.!)()',' ~~~~~•:sAdnMr>,tra!KJn decisions I · Likely e,areau States • Plans No Only accuracy The United ensus --- (

78228 • ACLU • APPENDIX 05 (DOCS 29-32) AL 3/23/19 188347

PX-123

To: Park-Su, Sahra (Federal) [REDACTED] Cc: Ron S Jarmin (CENSUS/ADEP FED) [[email protected]]; Enrique Lamas (CENSUS/ADDP FED) [[email protected]]; Kelley, Karen (Federal) [REDACTED]; Walsh, Michael (Federal) [REDACTED]; Lenihan, Brian (Federal) [REDACTED] From: Christa Jones (CENSUS/ADEP FED) Sent: Sat 2/24/2018 7:01:41 PM Importance: Normal Subject: Re: Draft Response to Question Received: Sat 2/24/2018 7:01:42 PM Sahra, I’m fine with this [REDACTED] On Feb 23, 2018, at 6:50 PM, Park-Su, Sahra (Federal) wrote Ron/Enrique/Christa, Thank you again for you all your assistance. Below is [REDACTED]. Please let us know if you have any questions, comments, or concerns. Have a great weekend. Sahra What was the process that was used in the past to get questions added to the decennial Census or do we have something similar where a precedent was established? [REDACTED] Sahra Park-Su

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 189348

Senior Policy Advisor Office of Policy and Strategic Planning U.S. Department of Commerce [REDACTED] Sent: Monday, January 29, 2018 9:18:03 AM From: John Maron Abowd (CENSUS/ADRM FED) To: Mark E Asiala (CENSUS/DSSD FED) Cc: Patrick J Cantwell (CENSUS/DSSD FED); Simson L Garfinkel (CENSUS/ADRM FED); Ron S Jarmin (CENSUS/ADEP FED); Enrique Lamas (CENSUS/ADDP FED); James Whitehorne (CENSUS/ADDC FED) Subject:Re: Questions from DoC on technical response to DoJ request I do not understand this: >The data were rounded according to the standard DRB rounding scheme for special tabs where zero estimates are presented as ‘0’, estimates ranging between one and seven are presented as ‘4’ = 0, and all estimates eight or higher are rounded to the nearest five (e.g., 10, 15, 20, ...). I think you mean: recode 0 to 0 recode 1-7 to 4 recode 8-12 to 10 recode 13-17 to 15 ...

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 190349

Please confirm. John M. Abowd, PhD Associate Director and Chief Scientist Research and Methodology U.S. Census Bureau Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media From: Mark E Asiala (CENSUS/DSSD FED) Sent: Monday, January 29, 2018 9:08:20 AM To: John Maron Abowd (CENSUS/ADRM FED) Cc: Patrick J Cantwell (CENSUS/DSSD FED); Simson L Garfinkel (CENSUS/ADRM FED); Ron S Jarmin (CENSUS/ADEP FED); Enrique Lamas (CENSUS/ADDP FED); James Whitehorne (CENSUS/ADDC FED) Subject: Re: Questions from DoC on technical response to DoJ request I confirmed with James Whitehorne that the following restrictions were put into place for the CVAP tabulation: 1. The disclosure rules included restrictions on which tables could be published at the tract and block-group level but there were no individual geography suppressions. Specifically at the tract and block group level, we only publish the tables

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 191350

of counts by race and ethnicity for the universes of citizens and citizens of voting age. The tables based on the universes of total population and total voting age population by race/ethnicity are not published at the tract and block-group level. This prevents the derivation of non-citizens and voting age non-citizens by race/ethnicity for these levels of geography. The data were rounded according to the standard ORB rounding scheme for special tabs where zero estimates are presented as ‘0’, estimates ranging between one and seven are presented as ‘4’ = 0, and all estimates eight or higher are rounded to the nearest five (e.g., 10, 15, 20, ...). 2. For special tabs, no data quality filtering based on reliability are enforced. Thus, buyers get what they ask for and we typically use the working relationship developed in the special tab request process to communicate the limitations of the data. 3. Zero given the information from #1 and #2 above. Mark E. Asiala Assistant Division Chief American Community Survey Statistical Design Decennial Statistical Studies Division U.S. Census Bureau Office: 301.763.3605 Room: 4K071-HQ E-mail: [email protected] census.gov Connect with us on Social Media

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 192351

From: John Maron Abowd (CENSUS/ADRM FED) Sent: Thursday, January 25, 2018 01:43 PM To: Mark E Asiala (CENSUS/DSSD FED) Cc: Patrick J Cantwell (CENSUS/DSSD FED); Simson L Garfinkel (CENSUS/ADRM FED); Ron S Jarmin (CENSUS/ADEP FED); Enrique Lamas (CENSUS/ADDP FED) Subject: Questions from Doc on technical response to DoJ request I was called by David Langdon, Chief Privacy Officer, DoC, with the following questions and requests vis-a-vis our technical response: 1. In the block-group CVAP tables, what are the disclosure avoidance suppression rules (with public parameters only)? 2. In the block-group CVAP tables, what MOE standards are used to determine publishability? 3. What percentage of the CVAP population data are suppressed for either confidentiality or MOE reasons? More questions are probably coming, but we should be able to get answer to these. Thanks, John John M. Abowd, PhD Associate Director and Chief Scientist Research and Methodology U.S. Census Bureau

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 193352

Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media From: Mark E Asiala (CENSUS/DSSD FED) [[email protected]] Sent: 1/29/2018 6:25:11 PM To: John Maron Abowd (CENSUS/ADRM FED) [[email protected]] CC: Patrick J Cantwell (CENSUS/DSSD FED) [[email protected]]; Simson L Garfinkel (CENSUS/ADRM FED) [[email protected]]; Ron S Jarmin (CENSUS/ADEP FED) [[email protected]]; Enrique Lamas (CENSUS/ADDP FED) [[email protected]]; James Whitehorne (CENSUS/ADDC FED) [[email protected]] Subject:Re: Questions from DoC on technical response to DoJ request That is correct. I see I forgot to delete some old text as I was editing which caused the confusion. Mark E. Asiala Assistant Division Chief American Community Survey Statistical Design Decennial Statistical Studies Division

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 194353

U.S. Census Bureau Office: 301.763.3605 Room : 4K071-HQ E-mail: [email protected] census.gov Connect with us on Social Media From: John Maron Abowd (CENSUS/ADRM FED) Sent: Monday, January 29, 2018 9:18:03 AM To: Mark E Asiala (CENSUS/DSSD FED) Cc: Patrick J Cantwell (CENSUS/DSSD FED); Simson L Garfinkel (CENSUS/ADRM FED); Ron S Jarmin (CENSUS/ADEP FED); Enrique Lamas (CENSUS/ADDP FED); James Whitehorne (CENSUS/ADDC FED) Subject: Re: Questions from DoC on technical response to DoJ request I do not understand this: >The data were rounded according to the standard DRB rounding scheme for special tabs where zero estimates are presented as ‘0’, estimates ranging between one and seven are presented as ‘4’ = 0, and all estimates eight or higher are rounded to the nearest five (e.g., 10, 15, 20, ...). I think you mean: recode 0 to 0 recode 1-7 to 4 recode 8-12 to 10

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 195354

recode 13-17 to 15 ... Please confirm. John M. Abowd, PhD Associate Director and Chief Scientist Research and Methodology U.S. Census Bureau Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media From: Mark E Asiala (CENSUS/DSSD FED) Sent: Monday, January 29, 2018 9:08:20 AM To: John Maron Abowd (CENSUS/ADRM FED) Cc: Patrick J Cantwell (CENSUS/DSSD FED); Simson L Garfinkel (CENSUS/ADRM FED); Ron S Jarmin (CENSUS/ADEP FED); Enrique Lamas (CENSUS/ADDP FED); James Whitehorne (CENSUS/ADDC FED) Subject: Re: Questions from DoC on technical response to DoJ request I confirmed with James Whitehorne that the following restrictions were put into place for the CVAP tabulation: 1. The disclosure rules included restrictions on which tables could be published at the tract and

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 196355

block-group level but there were no individual geography suppressions. Specifically at the tract and block group level, we only publish the tables of counts by race and ethnicity for the universes of citizens and citizens of voting age. The tables based on the universes of total population and total voting age population by race/ethnicity are not published at the tract and block-group level. This prevents the derivation of non-citizens and voting age non-citizens by race/ethnicity for these levels of geography. The data were rounded according to the standard ORB rounding scheme for special tabs where zero estimates are presented as ‘0’, estimates ranging between one and seven are presented as ‘4’ = 0, and all estimates eight or higher are rounded to the nearest five (e.g., 10, 15, 20, ...). 2. For special tabs, no data quality filtering based on reliability are enforced. Thus, buyers get what they ask for and we typically use the working relationship developed in the special tab request process to communicate the limitations of the data. 3. Zero given the information from #1 and #2 above. Mark E. Asiala Assistant Division Chief American Community Survey Statistical Design Decennial Statistical Studies Division U.S. Census Bureau Office: 301.763.3605 Room: 4K071-HQ E-mail: [email protected] census.gov Connect with us on Social Media

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 197356

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 198357

Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media *** From: John Maron Abowd (CENSUS/ADRM FED) [[email protected]] Sent: 1/29/2018 3:24:14 PM To: Ron S Jarmin (CENSUS/ADEP FED) [[email protected]]; Enrique Lamas (CENSUS/ADDP FED) [[email protected]] Subject: DoJ Technical Response Any guidance on when the next version needs to be ready? Thanks, John M. Abowd, PhD Associate Director and Chief Scientist Research and Methodology U.S. Census Bureau Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media *** From: John Maron Abowd (CENSUS/ADRM FED) [[email protected]] Sent: 1/29/2018 2:18:03 PM To: Mark E Asiala (CENSUS/DSSD FED) [[email protected]]

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 199358

CC: Patrick J Cantwell (CENSUS/DSSD FED) [[email protected]]; Simson L Garfinkel (CENSUS/ADRM FED) [[email protected]]; Ron S Jarmin (CENSUS/ADEP FED) [[email protected]]; Enrique Lamas (CENSUS/ADDP FED) [[email protected]]; James Whitehorne (CENSUS/ADDC FED) [[email protected]] Subject: Re: Questions from DoC on technical response to DoJ request I do not understand this: >The data were rounded according to the standard DRB rounding scheme for special tabs where zero estimates are presented as ‘0’, estimates ranging between one and seven are presented as ‘4’= 0, and all estimates eight or higher are rounded to the nearest five (e.g., 10, 15, 20, ...). I think you mean: recode 0 to 0 recode 1-7 to 4 recode 8-12 to 10 recode 13-17 to 15 ... Please confirm. John M. Abowd, PhD Associate Director and Chief Scientist Research and Methodology U.S. Census Bureau

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 200359

Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media From: Mark E Asiala (CENSUS/DSSD FED) Sent: Monday, January 29, 2018 9:08:20 AM To: John Maron Abowd (CENSUS/ADRM FED) Cc: Patrick J Cantwell (CENSUS/DSSD FED); Simson L Garfinkel (CENSUS/ADRM FED); Ron S Jarmin (CENSUS/ADEP FED); Enrique Lamas (CENSUS/ADDP FED); James Whitehorne (CENSUS/ADDC FED) Subject: Re: Questions from Doc on technical response to DoJ request I confirmed with James Whitehome that the following restrictions were put into place for the CVAP tabulation: 1. The disclosure rules included restrictions on which tables could be published at the tract and block-group level but there were no individual geography suppressions. Specifically at the tract and block group level, we only publish the tables of counts by race and ethnicity for the universes of citizens and citizens of voting age. The tables based on the universes of total population and total voting age population by race/ethnicity are not published at the tract and block- group level. This prevents the derivation of non-citizens and voting age non-citizens by race/ethnicity for these levels of geography.

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 201360

The data were rounded according to the standard ORB rounding scheme for special tabs where zero estimates are presented as ‘0’, estimates ranging between one and seven are presented as ‘4’= 0, and all estimates eight or higher are rounded to the nearest five (e.g., 10, 15, 20, ...). 2. For special tabs, no data quality filtering based on reliability are enforced. Thus, buyers get what they ask for and we typically use the working relationship developed in the special tab request process to communicate the limitations of the data. 3. Zero given the information from #1 and #2 above. Mark E. Asiala Assistant Division Chief American Community Survey Statistical Design Decennial Statistical Studies Division U.S. Census Bureau Office: 301.763.3605 Room: 4K071-HQ E-mail: [email protected] census.gov Connect with us on Social Media From: John Maron Abowd (CENSUS/ADRM FED) Sent: Thursday, January 25, 2018 01:43 PM To: Mark E Asiala (CENSUS/DSSD FED) Cc: Patrick J Cantwell (CENSUS/DSSD FED); Simson L Garfinkel (CENSUS/ADRM FED); Ron S Jarmin (CENSUS/ADEP FED); Enrique Lamas (CENSUS/ADDP FED) Subject: Questions from DoC on technical response to DoJ request

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 202361

I was called by David Langdon, Chief Privacy Officer, DoC, with the following questions and requests vis-a-vis our technical response: 1. In the block-group CVAP tables, what are the disclosure avoidance suppression rules (with public parameters only)? 2. In the block-group CVAP tables, what MOE standards are used to determine publishability? 3. What percentage of the CVAP population data are suppressed for either confidentiality or MOE reasons? More questions are probably coming, but we should be able to get answer to these. Thanks, John John M. Abowd, PhD Associate Director and Chief Scientist Research and Methodology U.S. Census Bureau Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media From: Mark E Asiala (CENSUS/DSSD FED) [[email protected]] Sent: 1/29/2018 2:08:20 PM To: John Maron Abowd (CENSUS/ADRM FED) [[email protected]] CC: Patrick J Cantwell (CENSUS/DSSD FED) [[email protected]]; Simson L Garfinkel (CENSUS/ADRM FED) [[email protected]];

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 203362

Ron S Jarmin (CENSUS/ADEP FED) [[email protected]]; Enrique Lamas (CENSUS/ADDP FED) [[email protected]]; James Whitehorne (CENSUS/ADDC FED) [James. [email protected]] Subject: Re: Questions from DoC on technical response to DoJ request I confirmed with James Whitehome that the following restrictions were put into place for the CVAP tabulation: 1. The disclosure rules included restrictions on which tables could be published at the tract and block-group level but there were no individual geography suppressions. Specifically at the tract and block group level, we only publish the tables of counts by race and ethnicity for the universes of citizens and citizens of voting age. The tables based on the universes of total population and total voting age population by race/ethnicity are not published at the tract and block- group level. This prevents the derivation of non-citizens and voting age non-citizens by race/ethnicity for these levels of geography. 2. The data were rounded according to the standard ORB rounding scheme for special tabs where zero estimates are presented as ‘0’, estimates ranging between one and seven are presented as ‘4’= 0, and all estimates eight or higher are rounded to the nearest five (e.g., 10, 15, 20, ...). 3. For special tabs, no data quality filtering based on reliability are enforced. Thus, buyers get what they ask for and we typically use the working

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 204363

relationship developed in the special tab request process to communicate the limitations of the data. 3. Zero given the information from #1 and #2 above. Mark E. Asiala Assistant Division Chief American Community Survey Statistical Design Decennial Statistical Studies Division U.S. Census Bureau Office: 301.763.3605 Room: 4K071-HQ E-mail: [email protected] census.gov Connect with us on Social Media From: John Maron Abowd (CENSUS/ADRM FED) Sent: Thursday, January 25, 2018 01:43 PM To: Mark E Asiala (CENSUS/DSSD FED) Cc: Patrick J Cantwell (CENSUS/DSSD FED); Simson L Garfinkel (CENSUS/ADRM FED); Ron S Jarmin (CENSUS/ADEP FED); Enrique Lamas (CENSUS/ADDP FED) Subject: Questions from DoC on technical response to DoJ request I was called by David Langdon, Chief Privacy Officer, DoC, with the following questions and requests vis-a-vis our technical response: 1. In the block-group CVAP tables, what are the disclosure avoidance suppression rules (with public parameters only)?

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 205364

2. In the block-group CVAP tables, what MOE standards are used to determine publishability? 3. What percentage of the CVAP population data are suppressed for either confidentiality or MOE reasons? More questions are probably coming, but we should be able to get answer to these. Thanks, John John M. Abowd, PhD Associate Director and Chief Scientist Research and Methodology U.S. Census Bureau Office 301.763.5880 (simulring on cell) Room 8H120 [email protected] census.gov Connect with us on Social Media

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 206365

Internal Document - Not for Public Release

September 20, 2017 MEMORANDUM FOR Associate Directorate for Research and Methodology (ADRM) From: Center for Survey Measurement (CSM) Subject: Respondent Confidentiality Concerns CSM researchers have noticed a recent increase in respondents spontaneously expressing concerns about confidentiality in some of our pretesting studies conducted in 2017. We recommend system- atically collecting data on this phenomenon, and development and pretesting of new messages to avoid increases in nonresponse among hard-to- count populations for the 2020 Census as well as other surveys like the American Community Survey (ACS). Below is a preview of findings relating to respondent confidentiality concerns from recent CSM projects, followed by a more detailed recommendation from CSM. These findings are drawn from usability interviews with English- and Spanish- speaking respondents (N=[REDACTED]), cognitive interviews with Spanish-speaking respondents (N=[REDACTED]), four focus groups with Spanish-speaking Field Representatives (FRs) (N= [REDACTED], five focus groups with Field Supervisors (FSs) and Field Representatives (N=[REDACTED]), and [REDACTED] focus groups with respondents (N [REDACTED]). These interviews and focus groups were conducted in different regions of the country in English, Spanish, Chinese, Korean, Vietnamese, Russian, and Arabic since

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 207366

January of 2017. All projects were small, qualita- tive studies and as such, unrepresentative of the population as a whole, and none of them were specifically designed to examine confidentiality concerns. However, respondents and field representatives spontaneously brought up these concerns at a much higher rate than CSM researchers have seen in previous pretesting projects, and as such, this information may have implications for nonresponse on U.S. Census Bureau studies and surveys. In particular, CSM researchers heard respondents express new concerns about topics like the “Muslim ban,” discomfort “registering” other household members by reporting their demographic characteristics, the dissolution of the “DACA” (Deferred Action for Childhood Arrival) program, repeated references to Immigration and Customs Enforcement (ICE), etc. FRs and FSs emphasized facing a “new phenomenon” in the field and reported that respondents’ fears, particularly among immigrant respondents, have increased markedly this year. Respondents reported being told by community leaders not to open the door without a warrant signed by a judge, and CSM researchers observed respondents falsifying names, dates of birth, and other information on household rosters. FRs requested additional training to help them overcome respondents’ fears regarding confidentiality and data sharing with other agencies like ICE, as well as materials they could share with respondents to reassure them about these concerns. Usability Findings (2017 PEGA Internet Self- Response Instrument; N = [REDACTED]

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 208367

Overall, [REDACTED] respondents who participated in usability interviews in the DC-metro area to pretest the 2017 PEGA internet self- response (ISR) instrument in English and Spanish intentionally provided incomplete or incorrect information about household members due to concerns regarding confidentiality, particularly relating to perceived negative attitudes toward immigrants. One Spanish-speaking respondent said she was uncomfortable “registering” other household members and tried to exit the survey at the dashboard when she realized she would have to provide information on others who live with her. She mentioned being afraid because of the current political climate and news reports about changing immigration policy. The researcher had to help the respondent delete the other household members from the roster to avoid a break-off; she only provided her own information. A second Spanish-speaking respondent filled out information about herself and three family members but intentionally left three or four roomers off the roster because, “This frightens me, given how the situation is now” and mentioned being worried because of their “[immigration] status.” Both Spanish-speaking respondents stated that they would not complete the survey at home. A third Spanish-speaking respondent, who the researcher had reason to believe was not concerned about whether his data would be shared with other federal agencies because of his status as legal resident in the country, commented:

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 209368

“Particularly with our current political climate, the Latino community will not sign up because they will think that Census will pass their information on and people can come looking for them.” This theme came up repeatedly even for those without concerns about the immigration status of members of their household. One English-speaking respondent entered false names and some incorrect dates of birth for his roommates because he was not comfortable providing their information without their consent due to data sharing concerns. A second English-speaking respondent did not report five unrelated household members (some of whom were immigrants) because she does not report their rental income to the IRS and because of what she referred to as the “Muslim ban.” It should be noted that this level of deliberate falsification of the household roster, and spontaneous mention of concerns regarding negative attitudes toward immigrants, is largely unprecedented in the usability interviews that CSM has been conducting since 2014 in preparation for the 2020 Ce nsus. In general, we assume that pretesting respondents are in fact more willing to fill out the survey than most respondents would be during the 2020 Census, given that they are being paid a cash incentive for their participation and being interviewed by a researcher with whom they have established rapport. As such, these concerns might be even more pronounced during a production survey than researchers observed during pretesting.

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 210369

Cognitive Findings (CBAMS Paper Testing; N [REDACTED].) Spanish-speaking respondents who participated in paper testing of the CBAMS (Census Barriers, Attitudes, and Motivators Survey) expressed concern about whether their answers might be shared with other government agencies. One respondent said, “The possibility that the Census could give my information to internal security and immigration could come and arrest me for not having documents terrifies me.” Later she commented that she was worried that her information could be used against her if she answered that she is not satisfied with the government here. She thought someone could say, ‘If you’re not satisfied, why are you here?’ and this could be used against her to expel her from the country. Respondent concerns on this survey were eye- opening for CSM researchers because some of the respondents who participated in cognitive interviews had previously taken part in CSM pretesting projects. Despite having participated in the past, they seemed visibly nervous and reticent and required extensive explanations regarding how their data would be used and their personal identifying information would be redacted. This behavior was in contrast to their demeanor during prior CSM pretesting projects. Multilingual Focus Groups on Doorstep Messages for the 2020 Census (N = [REDACTED]) Respondents also raised concerns in [REDACTED] focus groups conducted this spring in order to test doorstep messages that enumerators can use to

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 211370

overcome reluctance in the 2020 Census. These focus groups were conducted in English, Spanish, Chinese, Korean, Vietnamese, Russian, and Arabic, and the topic of confidentiality concerns came up in several groups. For example, Spanish-speakers brought up immigration raids, fear of government, and fear of deportation. Respondents talked about having received advice not to open the door if they fear a visit from Immigration and Customs Enforcement (ICE) and that they could instead ask that warrants be slipped under the door. They suggested that the Census Bureau have something in writing that enumerators could slip under the door to indicate why an enumerator is at a respondent’s home. They felt that the most important message to encourage participation was confidentiality and the greatest barriers to Latino participation are fear and mistrust. Several Chinese-speaking focus group respondents stated that the Chinese community’s main fear or concern was immigration status and how the data are used. They also expressed concern about opening the door to a government official and not wanting to be “investigated.” Arabic-speakers reported that they had concerns about their perception of the current environment as unwelcoming to Arabic-speaking immigrants and said that they feared deportation. One respondent said, “The immigrant is not going to trust the Census employee when they are continuously hearing a contradicting message from the media everyday threatening to deport immigrants.” Respondents wanted to have more

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 212371

assurance about how the data would be used before providing personal information. English-speakers expressed similar reservations when discussing the current “environment.” In one English focus group, respondents spontaneously expressed concerns that their personal information would be shared with other agencies, and mentioned in particular that data could be shared with Immigration and Customs Enforcement and the Department of Homeland Security. One participant recommended that Census materials should explicitly explain that personal information is not shared with these agencies. Overall, concerns about the confidentiality of data, including between agencies, negative perceptions of immigrants, and deportation emerged across languages in this project. Focus Groups with Spanish-speaking Field Representatives (N = [REDACTED].) CSM conducted four focus groups from July to September with Spanish-speaking Census Bureau Field Representatives who work in different states regarding the Spanish translation of a health survey. Many of the FRs spontaneously brought up the topic of an upsurge in respondent confidentiality concerns. Many FRs stated that before they can begin an interview, they have to spend several minutes calming respondents and gaining their trust due to the current “political state.” [REDACTED] said, “The politics have changed everything. Recently.” Another mentioned that this is especially relevant given that the DACA (Deferred Action for Child-

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 213372

hood Arrival) program is “on the chopping block.” FRs reported that some respondents worry about giving out legitimate names or completing the roster; they often do not feel comfortable giving out information about other people in the household. [REDACTED] said, “This may just be a sign of the times, but in the recent several months before anything begins, I’m being asked times over, does it make a difference if I’m not a citizen?” FRs reported that many Spanish-speaking respondents distrust the statement on confidentiality in the survey mailing materials, even when they understand it. Many respondents believe that “the less information they give out, the better. The safer they are.” [REDACTED] said that in June she was doing a Census Bureau survey interview with questions about citizenship status. A Spanish-speaking respondent answered that he was not a citizen, and then appeared to lie about his country of origin. When [REDACTED] started asking about his year of entry into the U.S., he “ shut down” and stopped responding to her questions. He then walked out and left her alone in the apartment, which had never happened to her during an interview before. [REDACTED] commented that she had seen this scenario many times while administering the ACS, although this was the first time she had heard of a respondent actually leaving the [REDACTED] alone in his or her home. She suggested that respondents might have concerns about confidentiality given “the current political climate.”

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 214373

A [REDACTED] added that she had observed Hispanic members of a household move out of a mobile home after she tried to interview them. She said, “There was a cluster of mobile homes, all Hispanic. I went to one and I left the information on the door. I could hear them inside. I did two more interviews, and when I came back, they were moving.... It’s because they were afraid of being deported.” FRs reported using various strategies to overcome respondents’ fears. They are often asked if they work for other federal agencies, and reassure respondents that this information is not reported to other federal agencies; their information is not shared with “immigration or taxes.” They explain that the respondent’s immigration status does not matter. The FRs reported that sometimes they encourage respondents to do the interview anonymously with fake names, when it seems like the respondent is about to refuse. The FRs recommended that ad campaigns be used to reduce the mistrust the public has toward completing our surveys. They also requested “an immigration letter” like one used on the NHANES (National Health and Nutrition Examination Survey) that mentioned “la migra” [a slang term for ICE] that was very effective. The FRs could use it selectively when it was needed. It clearly said that the Census Bureau was not in any way related with “la migra”. FRs were asked to share the most important change that they wanted to see made to the Spanish translation of the survey materials. In [REDACTED] focus group, the [REDACTED] FRs

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 215374

agreed unanimously that they would like an “immigration statement” to appear on mailing materials because of current “political issues.” They reported that immigration concerns are the “topic of the day” and that they always have to allay fears about immigration by saying, “We do not share information with other agencies.” They suggested that the statement should convey that while the Census Bureau is part of the federal government, it is a statistical agency, and that the respondent’s legal status in the country does not matter at all. Focus Groups with Field Supervisors and Field Representatives (N = [REDACTED] CSM conducted five focus groups in September with Field Supervisors and Field Representatives to collect feedback on FR training, the availability of printed materials in various languages, and the usage of printed materials during a recent housing survey operation. The topic of respondent concerns regarding confidentiality came up repeatedly in these focus groups. In [REDACTED] focus group of Field Supervisors, [REDACTED] reported having a respondent produce papers proving US citizenship of household members during an interview. [REDACTED] reported that each time she spoke to a Spanish- speaking respondent, her focus was on convincing the respondent of the confidentiality of their answers “given the political temperature these days.” One FS said, “we have to let [respondents] know where this information is going. That’s their biggest fear.” When asked if the training the FRs had received was adequate, [REDACTED]

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 216375

commented that more training was needed on respondent confidentiality concerns, but that “this climate didn’t exist before [when training was designed last time], when you did the study three years ago, so of course it wasn’t planned in there.” FSs reiterated that the main issue they saw was privacy concerns of Latino respondents, and that FRs should do more practice interviews where someone models those concerns and concerns about immigration so that the FRs are more prepared to respond adequately in the field. FRs who spoke a language other than Spanish or English (e.g., Cantonese) reported that completing interviews for the survey in question this year was much harder than the last time the survey was fielded: “Three years ago was so much easier to get respondents compared to now because of the government changes... and trust factors [and] also because of what happened here [in the United States]. . . Three years ago I didn’t have problems with the immigration questions.” [REDACTED] commented, “There will always be political situations that are out of our control .... Some- times I just come right out and say, this isn’t for immigration.” Even FRs who only speak English reported needing additional training for encountering households where respondents are especially fearful. [REDACTED] reported that respondents have been confusing him with someone from Immigration and Customs Enforcement (ICE, formerly known as INS). He reported that respondents that identified him as working for the government were hesitant to answer any questions,

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 217376

and it was difficult to gain their trust. [REDACTED] agreed that most incompletes were due to a distrust of the government. When asked whether their training adequately prepared them, [REDACTED] mentioned that training regarding concerns about ICE could not have been included in the training they received because it was a new phenomenon. The FRs in this focus group emphasized that they were having to reorder the questions in this housing survey to collect demographics last in order to avoid breakoffs. Spanish bilingual FRs shared many of the same concerns as the Field Supervisors, speakers of languages other than English or Spanish, and the monolingual English-speaking FRs. They emphasized that when completing interviews with Spanish-speaking households, immigration concerns were challenging and that respondents seemed fearful. They requested more training focusing on respondent fears, particularly immigrant respondents’ fears. They mentioned respondents giving out false names and reordering survey questions to collect demographics last. Recommendation Overall, these findings, in various languages from respondents, Field Representatives, and Field Supervisors across the country who have participated in recent projects are raising concerns within CSM regarding potential barriers to respondent participation in the 2020 Census, as well as other Census Bureau surveys. The findings listed above are a sampling of what CSM researchers have observed on recent projects, and these concerns were all expressed spontaneously

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 218377

to researchers during the course of pretesting various survey materials. These findings are particularly troubling given that they impact hard-to-count populations disproportionately, and have implications for data quality and nonresponse. A systematic pretesting study evaluating respondent confidentiality concerns, both from the perspective of respondents as well as Field Representatives, would shed light on the nature and prevalence of these concerns, particularly for Limited English Proficient (LEP) or immigrant populations in the U.S. Quantitative analysis could also be done to examine any changes in response rates, mode of administration, item nonresponse, or number of contact attempts for surveys such as the ACS among non-English speakers and hard-to-count, immigrant respondents. Similarly, we could review whether the number of residents reported or the number of unrelated household members within households has declined in recent months. In addition to gathering data on any uptick in confidentiality concerns that may exist, we recommend designing and pretesting wording that could address these concerns in mailing materials, the Decennial Internet Self Response instrument, FAQs provided to enumerators, etc. This text could inform respondents that the Census Bureau does not collect information on immigration status or religion (similar to the language stating that we do not collect social security numbers), or that we do not share data with agencies like ICE. Pretesting with respondents from a variety of

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 219378

backgrounds would be vital given that such a message could be reassuring to some respondents but may have other effects for different populations. Care should be taken in crafting new messages. CSM also recommends that additional training be provided to FRs across surveys regarding allaying respondents’ confidentiality concerns.

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 220379

*** Checklist for Disclosure Avoidance Officers 1. DRB Request No: CBDRB-2018-CDAR-014 2. Bypass: Yes 3. Name of Disclosure Avoidance Officer: William Wisniewski 4. Name of request: Memo to Department of Justice Request to Add Citizenship Question to the 2020 Census 5. Summary of request: The Department of Justice has requested block level citizen voting-age population estimates by OMB-approved race and ethnicity categories from the 2020 Census of Population and Housing. This request contains a memo in response to the DOJ, describing alternative approaches to obtaining this information. It contains national level estimates of citizenship statistics from Census, survey and administrative records. 6. Name of disclosure avoidance programmer: N/a, as all requested output was based on unweighted or weighted counts. Statistics in the memo were either rounded households, or percentages of individuals based on some demographic characteristics. A supporting file was prepared by David Brown, CES and used to ensure all underlying counts pass Census disclosure avoidance thresholds. 7. Date submitted to the Disclosure Review Board: N/a, DAO Bypass 8. Date reviewed by the Disclosure Review Board: N/a, DAO Bypass

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 221380

9. Date completed: 1/19/18 10. Location of data files/specs: On RM shared drive: [REDACTED] 11. Were all Disclosure Review Board rules /requirements followed? Note that if there are any questions about how to correctly perform the required disclosure avoidance procedures, please contact the Center for Disclosure Avoidance Research for help. Briefly summarize the disclosure avoidance procedures used to protect these data. Yes 12. After performing the disclosure avoidance procedures, how was the quality of the resulting data examined? For example, comparing cross tabulations of the data before and after disclosure avoidance, auditing programs for suppression, checks for inconsistencies in the data if certain values were altered. Data was not altered after I performed the disclosure review. 13. After completion, the checklist must be attached to a copy of the request and the Disclosure Review Board approval memo. The Disclosure Avoidance Officer must ensure that his or her division/program area maintains a copy of this docu mentat ion . Location of this information: This DAO checklist will be placed on the RM shared drive: [REDACTED] 14. Signature of Disclosure Avoidance Officer: (signed) William Wisniewski, 1/19/18

78228 • ACLU • 06 APPENDIX (DOCS 33 THRU 40) AL 3/22/19 222381

PX-140

Questions on the Jan 19 Draft Census Memo on the DoJ Citizenship Question Reinstate- ment Request 1. With respect to Alternatives B and C, what is the difference, if any, between the time when the data collected under each alternative would be available to the public? Since the collection of this data, whether from administrative records or from an enumerated question, occurs prior to the creation of the Microdata Detail File (MDF) from which all tabulations will be performed, there is no difference in the timing of when the data collected under either alternative B or C could be made available to the public. 2. What is the “2020 Census publication phase” (page 1 of the Detailed Analysis for Alternative B) versus Alternative C? Would there be any difference? The 2020 Census publication phase is a broad window stretching from the release of the apportionment counts by December 31, 2020 through the last data product or report published in FY 2023, the final year of decennial funding for the 2020 Census. However, as stated in the answer to question 1, this data could be made available to the public on the same schedule as any other post-apportionment tabulated data product regardless of whether alternative B or C is used in its collection.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 223382

3. What is the non-response rate for: (A) each question on the 2000 and 2010 Decennial Census short form and (B) each question on the 2010 ACS and most recent ACS? The table below shows the item non-response (INR) rate for each question on the 2000 and 2010 Decennial Census short form. This is the percentage of respondents who did not provide an answer to an item. Item Nonresponse Rates for 2000 and 2010 Short Form Person Questions Relationship Sex Age Hispanic Race Tenure Origin 2010 1.5 1.5 3.5 3.9 3.3 4.5 I 2000 I 1.3 I1.1 I 3.7 I 3.1 I 2.9 I 4.1 I Source: Rothhaas, Lestina and Hill (2012) Tables Notes and Soucre: Rothhaas, C., Lestina, F. and Hill, J. (2012) “2010 Decennial Census Item Nonresponse and Imputation Assessment Report” 2010 Census Program for Evaluations and Experiments, January 24, 2012. From report: The INR rate is essentially the proportion of missing responses before pre-editing or imputation procedures for a given item (i.e., the respondent did not provide an answer to the item). For INR, missing values are included in the rates, but inconsistent responses (i.e., incompatible with other responses) are considered non-missing responses.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 224383

Online link to 2010 report that has 2000 information as well. https://www.census.gov/ 2010census/pdf/2010_Census_INR_Imputation_ Assessment.pdf See attached spreadsheet for the non-response rates for the ACS. Note that these are internal use data. 4. What was the total survey response rate (i.e. percentage of complete questionnaires) for the 2000 long form and the 2000 short form? Of the incomplete long forms, what percentage left the citizenship question blank? Of the completed long forms, what percentage (if known) contained incorrect responses to the citizenship question? 5. For the 2000 long and short forms, what was the percentage unanswered (left blank) for each question (i.e., what percentage of the responses for each question (sex, race, ethnicity, income, citizenship, etc.) were left blank)? 6. What was the incorrect response rate for the citizenship question that was asked on the Long Form during the 2000 Decennial Census? Does the response rate on the 2000 Long Form differ from the incorrect response rate on the citizenship question for the ACS? 7. What is the incorrect response rate on other Decennial or ACS questions for which Census has administrative records available (for example, age, sex or income)?

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 225384

8. How does the Census presently handle responses on the (A) Decennial Census and (B) the ACS when administrative records available to the Census confirm that the response on the Decennial Census or ACS is incorrect? Is the present Census approach to incorrect responses based on practice/ policy or law (statute or regulation)? We have always based the short form Decennial Census and the ACS on self response, and while we have procedures in place to address duplicate or fraudulent responses, we do not check the accuracy of the answers provided to the specific questions on the Census questionnaire. This is a long established practice at the Census Bureau that has been thoroughly tested and in place since 1970, when the Census Bureau moved to a mail-out/respond approach to the Decennial Census. Title 13 of the U.S. Code allows the Census Bureau to use alternative data sources, like administrative records, for a variety of purposes, and we are using data in new ways in the 2020 Census. While this includes the use of administrative records data to fill in areas where a respondent does not provide an answer, we have not explored the possibility of checking or changing responses that a responding household has provided in response to the questionnaire. 9. Please explain the differences between the self-response rate analysis and the breakoff rate analysis. The range of breakoff rates between groups was far smaller than the range of self-response rates between groups.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 226385

10. The NRFU numbers are comparatively small – approximately one additional household for NRFU per Census enumerator. Is this really a significant source of concern? 11. Given that the breakoff rate difference was approximately 1 percent, why did Census choose to use the 5.1 percent number for assessing the cost of Alternative B? 12. Alternative C states that Census would use administrative data from the Social Security Administration, Internal Revenue Service, and “other federal and state sources.” What are the other sources? In addition to continuing the acquisition of the Social Security Administration and Internal Revenue Service data, the Census Bureau is in discussion with the U.S. Citizen and Immigra- tion Services (USCIS) staff to acquire additional citizenship data. 13. Is Census confident that administrative data will be able to be used to determine citizenship for all persons (e.g., not all citizens have social security numbers)? 14. For Alternative C, the memo says, “we assume the availability of these record linkage systems and associated administrative data” – does Census already have in place access to this data or would this need to be negotiated? If negotiated, for which data sets specifically?

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 227386

The Census Bureau has longstanding contractual relationships with the Social Security Administration and the Internal Revenue Service that authorize the use of data for this project. For new data acquired for this project (i.e., USCIS) we would estimate a six month development period to put a data acquisition agreement in place. That agreement would also include terms specifying the authorized use of data for this project. 15. Are there any privacy issues/sensitive information prohibitions that might prevent other agencies from providing such data? 16. How long would Census expect any negotiation for access to data take? How likely is it that negotiations would be successful? Are MOA’s needed/required? Current data available to the Census Bureau provide the quality and authority to use that are required to support this project. Additional information potentially available from USCIS would serve to supplement/validate those existing data. We are in early discussions with USCIS to develop a data acquisition agreement and at this time have no indications that this acquisition would not be successful. 17. What limitations would exist in working with other agencies like IRS, Homeland Security, etc. to share data? The context for sharing of data for this project is for a one-way sharing of data from these agencies to the Census Bureau. Secure file

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 228387

transfer protocols are in-place to ingest these data into our Title 13 protected systems. For those data already in-place at the Census Bureau to support this project, provisions for sharing included in the interagency agreement restrict the Census Bureau from sharing person-level microdata outside the Census Bureau’s Title 13 protections. Aggregates that have been processed through the Bureau’s disclosure avoidance procedures can be released for public use. 18. If Alternative C is selected, what is Census’s backup plan if the administrative data cannot be completely collected and utilized as proposed? 19. Does Census have any reason to believe that access to existing data sets would be curtailed if Alternative C is pursued? No we do not believe that any access to existing data sets would be curtailed if we pursue Alternative C. 20. Has the proposed Alternative C approach ever been tried before on other data collection projects, or is this an experimental approach? If this has been done before, what was the result and what were lessons learned? 21. Is using sample data and administrative records sufficient for DOJ’s request? 22. Under Alternative C, If Census is able to secure interagency agreements to provide needed data sets, do we know how long it would take to receive the data transmission

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 229388

from other agencies and the length of time to integrate all that data, or is that unknown? With the exception of the USCIS data, the data used for this project are already integrated into the 2020 Census production schema. In mid-to late 2018, we plan to acquire the USCIS data and with those data and our existing data begin to develop models and business rules to select citizenship status from the composite of sources and attach that characteristic to each U.S. person. We expect the development and refinement of this process to continue into 2019 and to be completed by third quarter calendar year 2019. 23. Cross referencing Census decennial responses with numerous governmental data sets stored in various databases with differing formats and storage qualities sounds like it could be complicated. Does Census have an algorithm in place to efficiently combine and cross reference such large quantities of data coming from many different sources? What cost is associated with Alternative C, and what technology/plan does Census have in place to execute? 24. For section C-1 of the memo, when did Census do the analyses of the incorrect response rates for non-citizen answers to the long form and ACS citizenship question? Were any of the analyses published? The comparisons of ACS, 2000 Decennial Census longform and SSA Numident citizen-

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 230389

ship were conducted in January 2018. This analysis has not been published. 25. Has Census corrected the incorrect responses it found when examining noncitizen responses? If not, why not? In the American Community Survey (ACS), and the short form Decennial Census, we do not change self-reported answers. The Decennial Census and the ACS are based on self-response and we accept the responses provided by households as they are given. While we have procedures in place to address duplicate or fraudulent responses, we do not check the accuracy of the answers provided to the specific questions on the Census questionnaires. This is a long established process at the Census Bureau that has been thoroughly tested and in place since 1970, when the Census Bureau moved to a mail-out/respond approach to the Decennial Census. 26. Has the Department of Justice ever been made aware of inaccurate reporting of ACS data on citizenship, so that they may take this into consideration when using the data? Not exactly. The Census Bureau is in close, regular contact with the Department of Justice (DOJ) regarding their data requirements. Our counterparts at DOJ have a solid understanding of survey methodology and the quality of survey data, and they are aware of the public documentation on sampling and accuracy surrounding the ACS. However, the specific rate of accuracy regarding responses to the

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 231390

ACS question on citizenship has never been discussed. 27. Why has the number of persons who cannot be linked increased from 2010 to 2016? 28. Independent of this memo, what action does Census plan to take in response to the analyses showing that non-citizens have been incorrectly responding to the citizenship question? The Census Bureau does not have plans to make any changes to procedures in the ACS. However, we will continue to conduct thorough evaluations and review of census and survey data. The ACS is focusing our research on the potential use of administrative records in the survey. For instance, we are exploring whether we can use IRS data on income to reduce the burden of asking questions on income on the ACS. We are concentrating initially on questions that are high burden, e.g., questions that are difficult to answer or questions that are seen as intrusive. 29. Did Census make recommendations the last time a question was added? Since the short form Decennial Census was established in 2010, the only requests for new questions we have received have been for the ACS. And, in fact, requests for questions prior to 2010 were usually related to the Decennial Census Long Form. We always work collaboratively with Federal agencies that request a new question or a change to a question. The first step is to review the data needs and the legal

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 232391

justification for the new question or requested changes. If, through this process, we determine that the request is justified, we work with the other agencies to test the question (cognitive testing and field testing). We also work collaboratively on the analysis of the results from the test which inform the final recommendation about whether or not to make changes or add the question. 30. Does not answering truthfully have a separate data standard than not participating at all? We’re not sure what you’re asking here. Please clarify the question. 31. What was the process that was used in the past to get questions added to the decennial Census or do we have something similar where a precedent was established? The Census Bureau follows a well-established process when adding or changing content on the census or ACS to ensure the data fulfill legal and regulatory requirements established by Congress. Adding a question or making a change to the Decennial Census or the ACS involves extensive testing, review, and evaluation. This process ensures the change is necessary and will produce quality, useful information for the nation. The Census Bureau and the Office of Manage- ment and Budget (OMB) have laid out a formal process for making content changes.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 233392

• First, federal agencies evaluate their data needs and propose additions or changes to current questions through OMB. • In order to be included, proposals must demonstrate a clear statutory or regulatory need for data at small geographies or for small populations. • Final proposed quest ions result from extensive cognitive and field testing to ensure they result in the proper data, with an integrity that meets the Census Bureau’s high standards. • This process includes several opportunities for public comment. • The final decision is made in consultation with OMB. • If approved, the Census Bureau implements the change. 32. Has another agency ever requested that a question be asked of the entire population in order to get block or individual level data? Not to our knowledge. However, it is worth pointing out that prior to 1980 the short form of the Decennial Census included more than just the 10 questions that have been on the short form since 1990. 33. Would Census linking of its internal data sets, with other data sets from places like IRS and Homeland Security, have an impact on participation as well (i.e. privacy concerns)?

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 234393

The potential that concerns about the use of administrative records could have an impact on participation has always been a concern of ours, and it’s a risk that we’re managing on our risk register. We’ve worked closely with the privacy community throughout the decade, and we established a working group on our National Advisory Committee to explore this issue. We’ve also regularly briefed the Congress about our plans. At this stage in the decade there does not appear to be extensive concerns among the general public about our approach to using administrative records in the Nonresponse Operation or otherwise. We will continue to monitor this issue. 34. Would Alternative C require any legislation? If so, what is the estimated time frame for approval of such legislation? 35. Census publications and old decennial surveys available on the Census website show that citizenship questions were frequently asked of the entire population in the past. Citizenship is also a question on the ACS. What was the justification provided for citizenship questions on the (A) short form, (B) long form, and (C) ACS? In 1940, the Census Bureau introduced the use of a short form to collect basic characteristics from all respondents, and a long form to collect more detailed questions from only a sample of respondents. Prior to 1940, census questions were asked of everyone, though in some cases only for those with certain characteristics. For

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 235394

example, in 1870, a citizenship question was asked, but only for respondents who were male and over the age of 21. We have never asked a question about citizenship on the short form. Comment [JMA(F1]: This is inconsistent with the paragraph above. I suggest, “Since moving to the short form in 1940, we have never ... Beginning in 2005, all the long-form questions – including a question on citizenship – were moved to the ACS. 2010 was the first time we conducted a short-form only census. The citizenship question is included in the ACS to fulfill the data requirements of the Department of Justice, as well as many other agencies including the Equal Employment Opportunities Commission, the Department of Health and Human Services, and the Social Security Administration.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 236395

PX-142

Question Assigned to Completed 1 Reist X 2 Reist X 3a Abowd: Mule X 3b Velkoff X 4 Velkoff X 5 Abowd: Mule X 6 Abowd: Brown X 7 Abowd: Brown X 8 Velkoff X 9 Abowd: Brown X 10 Abowd X 11 Abowd: Brown X 12 Abowd: Berning X 13 Abowd X 14 Abowd: Berning X 15 Abowd X 16 Abowd: Berning X 17 Abowd: Berning X 18 Abowd X 19 Abowd X 20 Abowd X 21 Abowd X

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 237396

23 Abowd: Berning X 23 Abowd: Mule X 24 Abowd: Brown X 25 Velkoff X 26 Reist X 27 Abowd: Brown X 28 Velkoff X 20 Velkoff X 30 Velkoff X 31 Velkoff X 32 Reist: Whitehorne X 33 Reist X 34 Abowd X 35 Reist: Dinwiddie X

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 238397

Missouri Johnson, Marcellina (Federal) From: Fidel, Matt Sent: Thursday, February 15, 2018 5:12 PM To: DOCExecSec Cc: Pesner, Rabbi Jonah; Weinstein, Barbara Subject: National Jewish Organizations Urge Secretary Ross to Reject the Department of Justice Request to Add a Citizenship Question to the 2020 Census Attachments: National Jewish Organizations Urge Secretary Ross to Reject the Department of Justice Request to Add a Question about Citizenship to the 2020 Census FINAL.docx February 15, 2018 The Honorable Wilbur Ross Secretary of Commerce U.S. Commerce Department 1401 Constitution Avenue NW Washington, DC 20230 Dear Secretary Ross: On behalf of the 10 undersigned Jewish organizations, we urge you to reject the Department of Justice’s harmful request to add a new citizenship question to the 2020 Census. This additional question is unnecessary and would fundamentally threaten the integrity and accuracy of the decennial census, with wide-ranging implications for our nation.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 239398

If the Census Bureau were to grant the Depart- ment of Justice’s request, it raises the likelihood of suppressing response rates from immigrant and other minority communities. From the ban on entry of immigrants from Muslim-majority countries to the termination of DACA, America’s immigrant communities feel increasingly vulnerable. A new Census question about citizenship will raise fears about such information now or in the future being used against them or their loved ones. This will potentially lower Census response rates and undermine the Census’s accuracy. Depressed Census participation would have far reaching consequences, as the data gathered by the Census is relied upon to allocate federal funding and determine congressional representation. If communities with large immigrant populations are undercounted by the Census, the government’s ability to meet the needs of the American people through the provision of essential services and aid dollars will be thwarted. Further, the interests of immigrant communities would not be accurately represented in Congress if the congressional apportionment process is based upon flawed data, undermining our representative democracy. The Justice Department stated that the addition of the citizenship question will facilitate enforcement the Voting Rights Act. However, the federal government continues to conduct the American Community Survey to obtain estimates of the citizen population, the data from which has been deemed suitable for use in Voting Rights Act enforcement cases. Since the inception of the decennial Census in 1790, it has counted citizens and non-citizens alike. It has not included

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 240399

questions about citizenship since 1960. Moreover, all questions that are included on the Census are carefully designed and tested to ensure that the data collected is accurate. Adding a question to the Census at this stage of the planning process would disrupt preparation and increase costs, in addition to threatening the accuracy of the data. Throughout history, the Jewish community has valued broad participation in civic life. Even in biblical times, Jewish leaders understood the importance of a fair and accurate census. The Torah tells us that in the wilderness of Sinai, God commanded Moses to take a head count of the people (Numbers 1:2). Our modern-day responsibility to support the engagement of all people in the life and well-being of our communities is no less significant. Historically, the Census has undercounted people of color and immigrants. We urge you not to compound this problem and, instead, protect the integrity of the 2020 Census by rejecting the Department of Justice’s request to add a question about citizenship. If you have any questions or wish to discuss this further, please contact Rabbi Jonah Pesner, Director of the Religious Action Center of Reform Judaism, at [email protected] or 202-387-2800. Sincerely, Anti-Defamation League Bend the Arc Jewish Action Central Conference of American Rabbis Hadassah, the Women’s Zionist Organization of America, Inc. Jewish Council for Public Affairs

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 241400

Jewish Federations of North America Jewish Women International MAZON: A Jewish Response to Hunger National Council of Jewish Women Union for Reform Judaism Matt Fidel Legislative Assistant Religious Action Center of Reform Judaism (202) 387-2800 | [email protected] twitter.com/TheRAC | facebook.com/TheRAC | instagram.com/theRACgram [LOGO]

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 242401

PX-607

To: Comstock, Earl (Federal) [PII] From: Uthmeier, James (Federal) Sent: Fri 8/11/2017 8:05:48 PM Importance: Normal Subject: Re: Census paper Received: Fri 8/11/2017 8:05:51 PM Census Memo Draft Aug 11 2017.docx Thanks Earl, clean copy attached. I can swing a call any time after 4:30 today. James From: Comstock, Earl (Federal) Sent: Friday, August 11, 2017 3:40 PM To: Uthmeier, James (Federal) Subject: Re: Census paper Thanks James. Please take a look at the attached edits. If you agree then we can send to the Secretary, who wanted to have a call today to discuss. Earl From: “Uthmeier, James (Federal)” [PII] Date: Friday, August 11, 2017 at 10:18 AM To: “Comstock, Earl (Federal)” [PII] Subject: Re: Census paper Made a couple small edits for clarity. Also, I have not yet sent this to Peter. Just let me know if you want me to loop him in—I think he is heading out pretty early today, and I’m tied up 11-1, but maybe we can walk through with him early next week.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 243402

From: Uthmeier, James (Federal) Sent: Friday, August 11, 2017 9:55:52 AM To: Comstock, Earl (Federal) Subject: Re: Census paper Earl- A draft, predecisional and privileged memo is attached. I know he likes short briefing materials, but I wanted to be more thorough given the issue and our uncertainty regarding the exact question(s) being presented. I will keep working to clean it up and am happy to incorporate any edits. I am out of the office for some MBDA and infrastructure meetings but can be reached on my cell. I’ll be able to talk today other than 11-1. Will be working over the next hour to clean this up a bit. If you want to provide some handwritten comments, you can deliver to Barb (OGC secretary) and she will get them to me quickly. I have some new ideas/recommendations on execution that I look forward to discussing. Ultimately, we do not make decisions on how the data should be used for apportionment, that is for Congress (or possibly the President) to decide. I think that’s our hook here. Best, James

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 244403

From: Comstock, Earl (Federal) Sent: Friday, August 11, 2017 8:11:41 AM To: Uthmeier, James (Federal) Subject: Re: Census paper Great. Thanks! Earl Sent from my iPhone >On Aug 11, 2017, at 7:45 AM, Uthmeier, James (Federal) <[PII]> wrote: > >Earl- > >Finishing this up this morning and will have a memo to you by 930. > >James > >Sent from my iPhone

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 245404

UNITED STATES DEPARTMENT OF COMMERCE Economics and Statistics Administration U.S. Census Bureau Washington, DC 20233-0001 [Logo] December 22, 2017 MEMORANDUM FOR Ron S. Jarmin Performing the Non- exclusive Functions and Duties of the Director From: John M. Abowd Chief Scientist and Associate Director for Research and Methodology Subject: Feasibility of Enhancing the PL94-171 Redistricting Data [This memorandum and the accompanying white paper contain no confidential data. The tables in the white paper and the estimates in this memo were cleared for release to the public under CBDRB- 2017-CDAR-001.] Summary Based on balanced consideration of multiple factors of quality, cost and feasibility, we recommend that the citizenship data for Department of Justice Voting Rights Act enforcement be obtained through the use of administrative records and not through the addition of a question to the decennial census instrument.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 246405

Citizenship, race, and ethnicity data for the voting- age population are essential to designing legislative districts that meet the criteria for nondiscrimination according to Section 2 of the Voting Rights Act. The Census Bureau currently supports this requirement with two distinct publications: the PL94-171 redistricting data (PL94), which must be released by April 1st of the year following a decennial census, and the Citizen Voting Age Population by Race and Ethnicity (CVAP) data, which are published annually in February using the most current 5-year American Community Survey (ACS) data. The Department of Justice and redistricting experts, partisan and bi-partisan, combine these data to produce estimates of the citizen voting age population by race and ethnicity at the lowest feasible level of geography, usually a census block. Neither the PL94 nor the CVAP tabulations contain estimates of the citizen voting age population by race and ethnicity at the block level. For PL94, this is because there is no citizen variable on the census questionnaire. For CVAP, this is because the 5-year ACS estimates do not go below the block-group level, and even there often have margins of error that make them difficult to use for creating block-level estimates in combination with PL94 via statistical methods. The direct solution to this problem is to make a citizenship variable available on the 2020 Census Edited File (CEF), the internal, confidential data file from which the PL94 tabulations are produced. If citizenship were available on that file, the PL94 tabulations could be restructured to include direct estimates of the citizen voting age population by race and ethnicity at the block level. These

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 247406

tabulations would have essentially the same accuracy as current PL94 and Summary File 1 (SF1) data. There are two alternative methods for accomplishing the addition of citizenship to the CEF. The first method is to ask the question on the 2020 Census, just as we currently do on the ACS and used to do on the decennial census long form. The second method is to load the citizenship variable onto CEF by record linkage using administrative data. There are advantages and disadvantages to both methods. The advantages of directly asking the question are (1) the provenance of the data is transparent and (2) the data are contemporaneous with the census by construction. The disadvantages are (1) potential negative impact on voluntary cooperation with the census, and (2) poorer quality citizenship data than would be available through administrative records. The advantages of using administrative records are better quality data than result from directly asking citizenship, and (2) cost savings to the census from avoiding the need to redesign questionnaires and increase nonresponse follow-up due to lower voluntary compliance. The disadvantages of using administrative data are (1) some risk of differential incomplete coverage due to incompleteness of these data for some foreign-born subpopulations, and additional processing complexity during the critical period between the closeout of the Decennial Response File (DRF), the end of data acquisition from the census operations, and the delivery of the Census Edited File.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 248407

Analysis We were not able to find any randomized controlled trials of the census or ACS questionnaires with and without the citizenship question. We conducted a limited analysis of the incremental field burden using the following natural experiment. Compare the first mailing response rates in the 2010 Census and the 2010 ACS for the same housing units. Response rates for citizens and noncitizen households are both lower in the ACS than in the 2010 Census, however the response rate for the noncitizen households falls by 5.1 percentage points more than the decline for citizens. Assuming that the number of households with at least one noncitizen is about 7,435,000 (+/– 47,000),1 this implies an incremental burden of 380,000 households in non-response follow-up. 380,000 is approximately 0.3% of the 2016 Population estimate of 117,700,000 households in the U.S. At $100,000,000 incremental NRFU cost per 1% decline in first mailing response rates, this translates to approximately $32,000,000 cost due to the decline in response rates from including the citizenship question. The cost of implementing our recommended solution has not been fully vetted. Accounting for ten cycles of processing between January 1, 2018 and April 1, 2020 and two senior staff FTEs to do the required modeling, the cost is less than $1,000,000. These estimates include the burden on the 2020 Census processing of the Census Unedited File.

1 2011 estimate from https://www.census.gov/prod/ 2013pubs/acsbr11-15.pdf. The 7,435,000 estimate is for households with a noncitizen head. The estimate for households with at least one noncitizen is necessarily greater.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 249408

We investigated the availability of directly reported citizenship data on all household surveys. The white paper concludes that there is insufficient data collected between 2000 and 2015 to use this source alone. It also concludes that the administrative record data are superior to the direct reports in several important dimensions. First, using historical direct reports creates problems with the timeliness of the citizenship status for naturalized citizens who acquired that status between the time that they responded and the reference date for the 2020 Census. Second, we document that there is good evidence that citizenship is accurately reported by citizens, but less accurately self-reported by household responders. This accuracy deficit in the self-responses may be due to the inherent difficulty of securing such information from the householder or proxies when they pertain to someone other than the respondent. The accuracy deficit may also be due to the sensitivity of the citizenship question itself. The white paper documents with preliminary evidence that acquiring citizenship status from administrative records is very likely to produce more accurate and timely data overall than asking the question directly, and then handling the item nonresponse in the edit and imputation phase of the 2020 Census. Nevertheless, we note that if the question is asked, the administrative data sources discussed in the white paper could be a valuable supplement to that phase. Recommendation The accompanying white paper proposes the creation of PL94 block-level data with citizen voting age population by race and ethnicity, in addition to the total population by race and

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 250409

ethnicity, using citizenship data that have been linked from (i) a collection of high quality national administrative data that the Census Bureau has already integrated into 2020 Census systems and (ii) data from the United States Citizen and Immigration Services (USCIS) that would be acquired by executing a Memorandum of Under- standing with USCIS. We would develop our own edit and imputation system for the citizenship variable. The tabulation variable would be added to CEF and available to the 2020 Disclosure Avoidance Subsystem (DAS) for inclusion in a modified version of the proposed P2 table “Race/Ethnicity for the Population Age 18 and Over” where “Population Age 18 and Over” would be replaced by “Citizen Population Age 18 and Over.” This revision would allow the use of the same disclosure avoidance methodology, state-of-the-art differential privacy, currently available for the 2018 End-to-End Test and the enhanced methods, integrated PL94 and SF1 protection, planned for the 2020 Census itself. This version of P2 would be the first PL94 data produced at the block-level with estimates of the citizen voting age population by race and ethnicity and with accuracy comparable to the accuracy of Pl “Total population.” The P1 and P2 tables would tabulate race and ethnicity in the same manner as currently proposed. Tables P42 “Group Quarters Population by Group Quarters Type” and H1 “Occupancy Status” would not be modified. The 2020 Census questionnaire would not be altered, and the field operations would not have to be expanded to compensate for the lower rate of voluntary compliance predicted for a census that asks the citizenship question directly.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 251410

PX-523

From: Wilbur Rossi [PII] Sent: 8/10/2017 7:38:25 PM To: Comstock, Earl (Federal) [PII] Subject: Re: Census Matter I would like to be briefed on Friday by phone. I probably will need an hour or so to study the memo first.we should be very careful, about everything, whether or not it is likely to end up in the SC. WLR Sent from my iPad > On Aug 9, 2017, at 10:24 AM, Comstock, Earl (Federal) [PII] wrote: > > PREDECISIONAL AND ATTORNEY-CLIENT PRIVILEGED > > Mr. Secretary – we are preparing a memo and full briefing for you on the citizenship question. The memo will be ready by Friday, and we can do the briefing whenever you are back in the office. Since this issue will go to the Supreme Court we need to be diligent in preparing the administrative record. > > Earl > > On 8/8/17, 1:20 PM, “Wilbur Ross” [PII] wrote: [NOT RESPONSIVE/DELIBERATIVE] [NOT RESPONSIVE/DELIBERATIVE] were you on the call this morning about Census? They seem dig in about not sling the citizenship question and that raises the question of where is

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 252411

the DoJ in their analysis? If they still have not come to a conclusion please let me know your contact person and I will call the AG. Wilbur Ross > > Sent from my iPhone > >> on Aug 8, 2017, at 10:52 AM, Comstock, Earl (Federal) [PII] wrote: >> >> [NOT RESPONSIVE/DELIBERATIVE]

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 253412

PX-008

From: Wilbur Ross [PII] Sent: 8/10/2017 7:38:25 PM To: Comstock, Earl (Federal) [PII] Subject: Re: Census Matter I would like to be briefed on Friday by phone. I probably will need an hour or so to study the memo first.we should be very careful,about everything, whether or not it is likely to end up in the SC. WLR Sent from my iPad > On Aug 9, 2017, at 10:24 AM, Comstock, Earl (Federal ) [PII] wrote: > > PREDECISIONAL AND ATTORNEY-CLIENT PRIVILEGED > > Mr. Secretary – we are preparing a memo and full briefing for you on the citizenship question. The memo will be ready by Friday, and we can do the briefing whenever you are back in the office. Since this issue will go to the Supreme Court we need to be diligent in preparing the administrative record. > > Earl > > on 8/8/17, 1:20 PM, “Wilbur Ross” [PII] wrote: [NOT RESPONSIVE/ DELIBERATIVE] [NOT RESPONSIVE/ DELIBERATIVE] were you on the call this morning about Census? They seem dig in about not sling the citizenship question and that raises the question of where is

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 254413

the DoJ in their analysis? If they still have not come to a conclusion please let me know your contact person and I will call the AG. Wilbur Ross > > Sent from my iPhone > >> On Aug 8, 2017, at 10:52 AM, Comstock, Earl (Federal) [PII] wrote: >> >> [NOT RESPONSIVE/ DELIBERATIVE] To: Teramoto, Wendy (Federal) [REDACTED]@doc.gov] From: Comstock, Earl (Federal) Sent: Sat 9/16/2017 11:33:38 AM Importance: Normal Subject: Calls with DoJ Received: Sat 9/16/2017 11:33:38 AM Morning Wendy– Here is the memo I gave SWLR regarding my discussions with DoJ. Earl *** September 8, 2017 To: Secretary Wilbur Ross Fr: Earl Comstock Re: Census Discussions with DoJ In early May Eric Branstad put me in touch with Mary Blanche Hankey as the White House liaison in the Department of Justice. Mary Blanche worked for AG Sessions in his Senate office, and came with him to the Department of Justice. We met in person

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 255414

to discuss the citizenship question. She said she would locate someone at the Department who could address the issue. A few days later she directed me to James McHenry in the Department of Justice. I spoke several times with James McHenry by phone, and after considering the matter further James said that Justice staff did not want to raise the question given the difficulties Justice was encountering in the press at the time (the whole Comey matter). James directed me to Gene Hamilton at the Department of Homeland Security. Gene and I had several phone calls to discuss the matter, and then Gene relayed that after discussion DHS really felt that it was best handled by the Department of Justice. At that point the conversation ceased and I asked James Uthmeier, who had by then joined the Department of Commerce Office of General Counsel, to look into the legal issues and how Commerce could add the question to the Census itself. September 8, 2017 To: Secretary Wilbur Ross Fr: Earl Comstock Re Census Discussions with DoJ In early May Eric Branstad put me in touch with Mary Blanche Hankey as the White House liaison in the Department of Justice. Mary Blanche worked for AG Sessions in his Senate office, and came with him to the Department of Justice. We met in person to discuss the citizenship question. She said she would locate someone at the Department who could address the issue. A few days later she directed me to James McHenry in the Department of Justice.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 256415

I spoke several times with James McHenry by phone, and after considering the matter further James said that Justice staff did not want to raise the question given the difficulties Justice was encountering in the press at the time (the whole Comey matter). James directed me to Gene Hamilton at the Department of Homeland Security. Gene and I had several phone calls to discuss the matter, and then Gene relayed that after discussion DHS really felt that it was best handled by the Department of Justice. At that point the conversation ceased and I asked James Uthmeier, who had by then joined the Department of Commerce Office of General Counsel, to look into the legal issues and how Commerce could add the question to the Census itself.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 257416

PX-014

To: Park-Su, Sahra (Federal) [REDACTED] Cc: Ron S Jarmin (CENSUS/ADEP FED) [[email protected]] Enrique Lamas (CENSUS/ADDP FED) [[email protected]] Kelley, Karen (Federal) [REDACTED]; Walsh, Michael (Federal) [REDACTED]; Lenihan, Brian (Federal) [REDACTED] From: Christa Jones (CENSUS/ADEP FED) Sent: Sat 2/24/2018 7:01:41 PM Importance: Normal Subject: Re: Draft Response to Question Received: Sat 2/24/2018 7:01:42 PM Sahra, I’m fine with this. (This is not to say there weren’t some improvements and presentation changes for the topics between l990-2000-2010 and planned for 2020. I just want us all to be clear that the questionnaires was not identical from 1990 to now.) On Feb 23, 2018 , at 6:50 PM, Park-Su, Sahra (Federal) [REDACTED] wrote: Ron/Enrique/Christa, Thank you again for you all your assistance. Below is a draft response worked with Deputy GC Walsh. Please let us know if you have any questions, comments, or concerns. Have a great weekend. Sahra What was the process that was used in the past to get questions added to the decennial Census

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 258417

or do we have something similar where a precedent was established? No new questions were added to the 2010 Decennial Census, so there is no recent precedent for considering a request to add questions to a decennial census. Consistent with longstanding practice for adding new questions to the ACS survey, the Census Bureau is working with relevant stakeholders to ensure that legal and regulatory requirements are fulfilled and that the question would produce quality, useful information for the nation. As you are aware, that process is ongoing. Upon its conclusion, you will have all of the relevant data at your disposal to make an informed decision about the pending request from the Department of Justice. Sahra Park-Su Senior Policy Advisor Office of Policy and Strategic Planning U.S. Department of Commerce [REDACTED]

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 259418

PX-614

To: Kelley, Karen (Federal) [PII]; Hernandez, Israel (Federal) [PII] From: Park-Su, Sahra Sent: Mon 9/11/2017 10:39:25 PM Importance: Normal Subject: Re: Census Matter Follow-Up Received: Mon 9/11/2017 10:39:26 PM FYI Sent from my iPhone On Sep 11, 2017, at 6:38 PM, Uthmeier, James (Federal) [PII] wrote: Sahra– Here is the attached from last week. Apologies, I will make sure to keep you in the loop on this stuff. James From: Uthmeier, James (Federal) Sent: Friday, September 08, 2017 6:07 PM To: Comstock, Earl (Federal) [PII] Davidson, Peter (Federal) [PII] Cc: Teramoto, Wendy (Federal [PII] Subject: Re: Census Matter Follow up Earl– Here is an update on our efforts: [ACP/DELIBERATIVE] [ACP/DELIBERATIVE] Peter and I plan to continue discussing this on Monday and will look forward to reviewing with you and the Secretary. I will also continue to research the [ACP/DELIBERATIVE] throughout

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 260419

the weekend. If other thoughts or needs come up over the weekend, please do not hesitate to call me on my cell. Have a nice weekend, James From: Comstock, Earl (Federal) Sent: Thursday, September 07, 2017 6:13 PM To: Davidson, Peter (Federal) [PII] Uthmeier, James (Federal) [PII] Cc: Teramoto, Wendy (Federal) [PII] Subject: Re: Census Matter Follow-Up I suggest setting up a call for tomorrow. The Secretary is asking for progress on this. Earl From: “Davidson, Peter (Federal)” [PII] Date: Thursday, September 7, 2017 at 5:30 PM To: “Uthmeier, James (Federal)” [PII] “Comstock, Earl (Federal)” [PII]· Cc: Wendy Teramoto [REDACTED] Subject: RE: Census Matter·Follow-Up· I don’t believe it was the AG of Kansas, but instead the Sec of State that he was referring to. He is the state official that is heading up the anti-fraud working group. I am concerned about contacting him [ACP/DELIBERATIVE] contacted him, and the Sec will be is set up a meeting with some trusted before we do anything externally. I Mark Neumann, Ken and/or good political judgment. We timeout on any external contacts on this looking into this.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 261420

From: Uthmeier, James (Federal) Sent: Thursday, September 07, 2017 4:58 PM To: Comstock,, Earl (Federal) [PII] Cc: Davidson, Peter (Federal) [PII] Subject: Re: Census Matter Follow up Hi Earl– [ACP/DELIBERATIVE] [ACP/DELIBERATIVE] I’ll also spend more time looking at the [ACP/DELIBERATIVE] James On Sep 7, 2017, at 4:53 PM, Comstock, Earl (Federal) [PII] wrote: Hi Peter and James– As I discussed with James a little while ago, the Secretary would like an update on progress since the discussion yesterday regarding the citizenship question. If we could get a short email or memo today that would be great. Thanks. Earl

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 262421

Joint Stipulations in State of New York, et al. v. U.S. Dep’t of Commerce, et al, 18-CV-2921 (JMF)

I. Past Census Bureau Practices With Regard to Citizenship Information

a. Decennial Census Overview 1. The U.S. Constitution requires the federal government to conduct a Decennial Census counting the total number of “persons”—with no reference to citizenship status—residing in each state. 2. The Constitution provides that Representatives “shall be apportioned among the several States . . . according to their respective Numbers”; which requires “counting the whole number of persons in each State.” 3. The Constitution requires that this count be an “actual Enumeration” conducted every ten years. 4. Through the Census Act, Congress assigned the responsibility of making this enumeration to the Secretary of Commerce. 5. The Secretary of Commerce is charged with the responsibility to take a Decennial Census to create an actual enumeration of the United States population. 6. The central constitutional purpose of the Census Bureau in taking the Decennial Census is to conduct an enumeration of the total population. 7. To enable a person-by-person count, the Census Bureau sends a questionnaire to virtually every housing unit in the United States, which is directed

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 263422

to every person living in the United States and all persons living in the United States are legally required to respond. 8. If the Census Bureau does not receive a response to the questionnaire it then sends a Census Bureau staffer known as an enumerator to the housing unit to attempt to conduct an in person interview in order to collect the data. This process is the Non Response Follow Up (“NRFU”) operation. 9. If the initial NRFU visit does not result in collecting complete data for a household, administrative records may be used to enumerate a limited number of those households for which there is high quality administrative data about the household. 10. For those households without high-quality administrative records, an enumerator will attempt to re-contact the household in person. 11. If a third attempt to contact a household does not yield a response, a case will become “proxy- eligible.” 12. A proxy is someone who is not a member of the household—such as a neighbor, landlord, Postal worker, or other knowledgeable person who can provide information about the unit and the people who live there. 13. For a proxy-eligible case an enumerator will attempt three proxies after each re-contact attempt that does not result in an interview. 14. After the NRFU process is completed, the Census Bureau then counts the responses from every household, including those completed

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 264423

through the NRFU process, to determine the population count in each state. 15. Data from the Decennial Census are reported down to the census block level. 16. The population data collected through the Decennial Census determines the apportionment of seats in the U.S. House of Representatives among the states. 17. The population data collected through the Decennial Census also determines the number of electoral votes each state has in the Electoral College. 18. States also use Decennial Census data to draw congressional, state, and local legislative districts. 19. The federal government also uses Decennial Census data to allocate hundreds of billions of dollars in public funding each year, including to states and local governments. 20. Approximately 132 programs used Census Bureau data to distribute hundreds of billions of dollars in funds during fiscal year 2015. 21. Some demographic groups have proven more difficult to count in the Decennial Census than others. The Census Bureau refers to these groups as “hard-to-count.” 22. Racial and ethnic minorities, immigrant populations, and non-English speakers have historically been some of the hardest groups to count accurately in the Decennial Census. 23. Individuals identifying as Hispanic were undercounted by almost 5% in the 1990 Decennial Census.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 265424

24. The 2010 Decennial Census undercounted on net more than 1.5 million Hispanic and African American individuals. 25. The Census Bureau describes the undercounting of a particular racial and ethnic group in comparison to the overall net undercount or overcount of the population as a whole as a “differential undercount,” as distinct from a “net undercount” of the entire population. 26. The Census Bureau has developed a range of strategies to address the differential undercount of “hard-to-count” populations—including targeted marketing and outreach efforts, partnerships with community organizations, deployment of field staff to follow up with individuals who do not respond, and retention of staff with foreign language skills. 27. In the 2000 and 2010 Decennial Censuses, the Census Bureau designed and implemented public advertising campaigns to reach hard-to-count immigrant communities, including using paid media in over a dozen different languages to improve responsiveness. 28. For the 2000 and 2010 Decennial Censuses, the Census Bureau also partnered with local busi- nesses, faith-based groups, community organizations, elected officials, and ethnic organizations to reach these communities and improve the accuracy of the count.

b. The Long Form and the American Community Survey 29. The 1950 Decennial Census asked for the individual’s place of birth, and whether a foreign- born individual had been naturalized.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 266425

30. A question concerning citizenship did not appear on the Decennial Census questionnaire sent to every household in the United States (commonly referred to as the “short form”) in 1970, 1980, 1990, 2000, or 2010. 31. From at least the 1970 Decennial Census through the 2000 Decennial Census, in lieu of the short-form questionnaire the Census Bureau sent a long form questionnaire to approximately one in six households. 32. For the 1970 Decennial Census, the long form questionnaire, which contained additional questions, was sent to approximately one in five households. 33. For the 2000 Decennial Census, the long form questionnaire, which contained additional questions, was sent to approximately one in six households. 34. Data collected from the sample households surveyed with the long form were used to generate statistical estimates. 35. In the 1970, 1980, 1990, and 2000 Decennial Censuses, the long form Decennial Census questionnaire contained a question about citizenship status. 36. In the 1990 and 2000 Decennial Censuses, the citizenship status question on the long form questionnaire was preceded by a question about place of birth. 37. After the 2000 Decennial Census, the long form questionnaire was replaced by the American Community Survey.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 267426

38. After the 2000 Decennial Census, the functions performed by the long form have been replaced by the American Community Survey (“ACS”). 39. The ACS began operating in 2000 and was at full sample size for housing units in 2005, and for group quarters in 2006. 40. The ACS is a yearly survey of approximately 2% of households—about 3.5 million—across the United States. 41. A question concerning citizenship status currently appears as among one of more than 50 questions on the 28-page ACS questionnaire. 42. The citizenship status question on the ACS is preceded by a question asking where the person was born. 43. The citizenship question that appears on the ACS is not a binary yes/no question. The ACS citizenship question, asks whether the person was born in the United States, a U.S. territory, or abroad. 44. The data collected by the ACS allows the Census Bureau to produce estimates of Citizen Voting Age Population (“CVAP”). 45. CVAP data based on responses to the ACS are reported by the Census Bureau down to the census block group level. 46. Margins of error are reported with the ACS estimates and provide a measure of the sampling error associated with each estimate. 47. The ACS is intended to provide information on characteristics of the population, and the social and economic needs of communities.

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 268427

48. Unlike the Decennial Census, the ACS is not a complete enumeration, but rather a sample survey that is used to generate statistical estimates. 49. Because ACS estimates are statistical estimates based on a sample, the tabulations are weighted to reflect sampling probabilities and eligibility for NRFU, as well as well as to control to official population totals as established by the Population Estimates program. 50. Because the ACS collects information from only a small sample of the population, it produces annual estimates only for “census tract[s]” and “census-block groups.” 51. Although the ACS survey is conducted annually, ACS data from individual years can also be aggregated to produce multi-year estimates (commonly referred to as “1-year”, “3-year” or “5- year” estimates depending on the number of years aggregated together). 52. Multi-year ACS estimates have larger sample sizes than 1-year ACS estimates. Cumulating the five-year pooled estimates yields approximately a one-in–every-eight- household sample. 53. Multi-year ACS estimates have greater levels of statistical precision for estimates concerning smaller geographical units. 54. 1-year ACS estimates produce “[d]ata for areas with populations of 65,000+”; 1-year supplemental ACS estimates produce “[d]ata for areas with populations of 20,000+”, 3-year ACS estimates produced “[d]ata for areas with populations of 20,000+” until they were discontinued after the 2011-2013 3-year estimates, and 5-year ACS estimates produce “[d]ata for all areas.”

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 269428

55. The 2000 Decennial Census short form questionnaire did not include a question on citizenship. 56. The 2010 Decennial Census questionnaire did not include a question on citizenship.

II. The 2020 Decennial Census 57. A planned question on the 2020 Decennial Census questionnaire asks, “Is this person a citizen of the United States?,” with the answer options “Yes, born in the United States”; “Yes, born in Puerto Rico, Guam, the U.S. Virgin Islands, or Northern Marianas”; “Yes, born abroad of U.S. citizen parent or parents”; “Yes, U.S. citizen by naturalization – Print year of naturalization”; and “No, not a U.S. citizen”. 58. The 2020 Decennial Census questionnaire will pose questions including questions regarding sex, Hispanic origin, race, and relationship status. 59. A planned question on the 2020 Decennial Census questionnaire asks “Is this person of Hispanic, Latino, or Spanish origin?” 60. A planned question on the 2020 Decennial Census questionnaire asks “What is this person’s race?” 61. A planned question on the 2020 Decennial Census questionnaire asks how each person in the household is related to the person filling out the questionnaire. 62. A planned question on the 2020 Decennial Census questionnaire asks, “What is this person’s sex?”

78228 • ACLU • 07 APPENDIX (DOCS 42 THRU 52) AL 3/23/19 270429

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

Civil Action No. 1:18-cv-05025-JMF ______

NEW YORK IMMIGRATION COALITION, CASA DE MARYLAND, AMERICAN-ARAB ANTI-DISCRIMINATION COMMITTEE, ADC RESEARCH INSTITUTE, and MAKE THE ROAD NEW YORK, Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE; and WILBUR L. ROSS, JR., in his official capacity as Secretary of Commerce, and

BUREAU OF THE CENSUS, an agency within the United States Department of Commerce; and RON S. JARMIN, in his capacity as performing the non-exclusive functions and duties of the Director of the U.S. Census Bureau, Defendants. ______

DECLARATION OF STEVEN K. CHOI

Steven K. Choi, pursuant to the provisions of 28 U.S.C. § 1746, declares under penalty of perjury as follows:

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 271430

1. I am the Executive Director of the New York Immigration Coalition (“NYIC”). In that capacity, I am responsible in part for NYIC’s education and outreach efforts around the 2020 Census. I am also one of the NYIC executives responsible for the organization’s budgeting, fundraising, and policy priorities. I have been Executive Director of NYIC for over five years. 2. NYIC is an umbrella policy and advocacy organization for nearly 200 groups in New York State, representing the collective interests of New York’s diverse immigrant communities and organizations. NYIC is headquartered at 131 West 33rd St, New York, NY 10001. 3. NYIC’s mission is to unite immigrants, members, and allies so that all New Yorkers can thrive. NYIC envisions a New York State that is stronger because all people are welcome, treated fairly, and given the chance to pursue their dreams. NYIC pursues solutions to advance the interests of New York’s diverse immigrant communities and advocates for laws, policies, and programs that lead to justice and opportunity for all immigrant groups. It seeks to build the power of immigrants and the organizations that serve them to ensure their sustainability, improve people’s lives, and strengthen New York State. 4. NYIC’s nearly 200 members are dues-paying nonprofit organizations that are committed to advancing work on immigrant justice, empowerment, and integration. NYIC’s member organizations— located throughout New York State and beyond— all share NYIC’s mission to serve and empower immigrant communities. NYIC’s members include

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 272431

grassroots community groups, social services providers, large-scale labor and academic institutions, and organizations working in economic, social, and racial justice. Representatives of NYIC’s member organizations serve on the NYIC Board of Directors. 5. The Decennial Census is a critical and constitutionally-mandated data-gathering instrument, used to distribute hundreds of billions of dollars in federal resources and to apportion political power at the federal, state, and local levels. The importance of a complete and accurate Decennial Census is significant and requires a direct inquiry of every person in the United States. 6. As such, NYIC, its member organizations, and the communities we serve all have a fundamental interest in ensuring as complete and accurate a Decennial Census as possible. Among other Census-guided programs, NYIC member organizations receive funding through the Medical Assistance Program, also known as Medicaid; the Special Supplemental Nutrition Program for Women, Infants, and Children; the State Children’s Health Insurance Program; programs authorized under the Workforce Innovation and Opportunity Act; English Language Acquisition Grants; the Corporation for Community & National Service, which operates the AmeriCorps program; and formula grants authorized by the Violence Against Women Act. 7. NYIC also understands that data from the Decennial Census provides the basis for apportioning political representation at the federal, state, and local levels. An undercount for immigrant

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 273432

communities of color under the Decennial Census would unduly harm our members’ level of political representation at all levels of government. 8. Because a complete and accurate count is critical to ensuring that our member organizations and the communities they serve receive the government funding and full political representation to which they are entitled, NYIC has an ongoing commitment to promoting engagement in the Decennial Census among individuals served by its member organizations. 9. During the 2010 Census cycle, NYIC partnered with the New York Community Media Alliance to launch an outreach campaign to boost immigrant participation in the Census. As part of that effort, NYIC coordinated public service announcements in 24 languages that appeared in 69 newspapers. NYIC also held press briefings with elected officials. These efforts helped to increase New York City’s mail-in 2010 Census participation rate by approximately 3%. 10. For the 2020 Census, NYIC has already begun its outreach efforts. Since the beginning of 2018, it has helped form New York Counts 2020, a growing, non-partisan coalition of more than 50 diverse organizational stakeholders across New York to advocate for a fair and complete enumeration. This broad-based coalition, which was formally launched in March 2018, is composed of racial, ethnic, immigrant, religious, health, education, labor, housing, social services, and business groups working in partnership with state and local government officials.

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 274433

11. NYIC is investing resources to solidify the work and reach of New York Counts 2020 through robust advocacy, outreach, and mass educational forums. It has already begun disseminating online petitions, petitioning Community Boards to pass resolutions for a fair and accurate count, and co- convened an all-day statewide conference, “Making New York Count in 2020.” NYIC will continue coordinating the working committees of New York Counts 2020, including by: coordinating “train the trainer” sessions throughout the state to equip leaders with tools to educate their communities on the importance of the Census; devising effective messaging to convince hard-to-reach communities to participate; empowering coalition members to assist their communities in completing the Census online; and advocating to ensure that there are no unnecessary barriers impeding marginalized communities from being counted while also ensuring their privacy is protected. 12. NYIC has been and remains committed to Census education and outreach work in part because NYIC understands that immigrants and communities of color have been historically undercounted by the Census. From our work in the community, we understand that one reason that immigrants and communities of color have been undercounted is a distrust of government officials and a fear of turning over personal information to the government. 13. This level of fear and distrust of government among immigrants and communities of color has been exacerbated by the Trump Administration and its officials’ hostility to these communities, as

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 275434

demonstrated through numerous acts and statements. Immigrant communities of color, which historically have been reluctant to engage with government officials, are even more reluctant now due to the consistent racism and xenophobia exhibited by the Administration and its officials. Among the racist and xenophobic acts that the Trump Administration has undertaken include banning individuals from six majority Arab and/or Muslim countries from entering the United States; rescinding the Deferred Action for Childhood Arrivals (“DACA”) program, which allowed 800,000 individuals—90% of whom are Latino—brought to this country as children to legally reside and work in the United States; rescinding Temporary Protected Status programs for individuals from El Salvador, Honduras, Nicaragua, Haiti, and Nepal; calling for an end to the diversity visa lottery, a program in which over 40% of individuals admitted are from Africa, while another 30% are from Asia; and proposing to end family-based immigration, which would disproportionately harm immigrants from Latin America and Asia. NYIC has consistently fought these efforts to intimidate and marginalize immigrants of color. 14. Now, New York immigrant communities’ heightened fear of interacting with government workers has increased even further due to the decision to add the citizenship question. The citizenship question creates an incremental obstacle to Census participation because it ties immigrant communities of color’s fear directly to the Decennial Census instrument. By adding a citizenship question to the Decennial Census, the Trump Administration has taken advantage of a unique

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 276435

opportunity to bring their campaign to intimidate and marginalize immigrants into the homes of every immigrant. The citizenship question threatens to put all immigrant respondents, as well as their families, loved ones, and neighbors, in a bind: Identify your disfavored status to a hostile administration or risk the loss of critical federal resources and political power. For an administration that has found myriad ways to threaten and disparage immigrants, the citizenship question presents a singularly intrusive and effective method of attacking immigrants—one that has generated an incremental and heightened fear for immigrant communities of color. 15. In its already extensive 2020 Census outreach, NYIC has faced, and will continue to face, a more difficult Census-response environment due to New York immigrant communities’ heightened fear of interacting with government workers because of the addition of the citizenship question. This fear extends not only to undocumented immigrants or non-citizens with legal status, but also to family and household members of non-citizens who will be concerned that participating might endanger their loved ones. 16. The decision to add a citizenship question to the 2020 Decennial has required NYIC to make substantial and additional investments to achieve Census participation rates comparable to what we what would have achieved absent this decision. Prior to the addition of the citizenship question, NYIC had planned to spend approximately $625,000 on Census education and outreach over a three-year period ahead of the 2020 Census, with

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 277436

most the spending planned for 2019 and 2020. However, to the best of my knowledge, over the next three years, NYIC is planning to spend approximately $1 million on community education and outreach efforts to work towards a complete and accurate count within the communities that NYIC and its member organizations serve— representing an increase of approximately 60% over what the organization would have spent in the absence of a citizenship question. 17. So far in 2018, as a result of the decision to add a citizenship question, NYIC has spent at least $93,000 on Census-related activities that it would have not spent otherwise. NYIC anticipates spending an additional amount in excess of $282,000 between now and May 2020 as a result of the citizenship question. 18. To address the fear and confusion caused by the citizenship question among the immigrant communities that NYIC and its members serve, NYIC has had to begin its Census education and outreach efforts substantially earlier and engage many more staff, members, and partners than planned. Prior to the decision to add a citizenship question, NYIC did not anticipate having to commence significant Census education and outreach work until the Summer of 2019. Instead, as a result of the announcement of Secretary Ross’ decision, NYIC had to accelerate the start of significant Census work to March 2018. 19. The fear and confusion brought on by the decision to add a citizenship question prompted NYIC to hire a dedicated, full-time senior census fellow at a cost of approximately $36,000 in the

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 278437

year to date, whose responsibilities included accelerating the launch of New York Counts 2020 and organizing a conference that was prompted in large part by the addition of the citizenship question. Absent the citizenship question, NYIC would not have hired a dedicated, full-time senior census fellow until 2019. NYIC also hired multiple paid Census Interns to engage with NYIC’s ongoing communications, training, and education needs as they related to the citizenship question, at a cost of nearly $10,000 in the year to date. Without the citizenship question, NYIC would not have hired any Census-focused interns, but now anticipates spending $50,000 on interns through May 2020. NYIC also expended resources to broaden the reach of the New York Counts 2020 conference to address concerns among NYIC’s members and their communities arising out of the decision to add a citizenship question, which cost over $19,000— exclusive the cost of the time for organizational staff or the senior census fellow. NYIC has diverted and anticipates continuing to have to divert 10% of the staff time from the organization’s managers of member engagement for the Long Island, Western New York, Hudson Valley, and Central New York regions to address concerns related to the citizenship question, costing approximately $19,000. In 2019, NYIC expects to divert approximately 50% of the staff time for these four positions to Census work as a result of the citizenship question. Since March 2018, NYIC has also had to divert approximately 20% of staff time from the organization’s director of immigration policy to staff work, at a cost of approximately $14,000 to the organization. NYIC is also in the process of hiring a full-time

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 279438

manager of democracy policy to support the NYIC’s Census policy work, especially as relates to policy issues generated by a potential citizenship question, including Title 13 privacy protections, what an undercount due to the citizenship question would mean for immigrant and refugee New Yorkers and the communities they live in, and data concerns. We anticipate 40% of the Manager’s time to be spent on work generated as a result of the citizenship question through May 2020, costing approximately $41,000. NYIC has also made expenditures and anticipates continuing to have to make expenditures for overhead and benefits for each of these positions that together account for approximately one-and-a-half times the amount of salary paid for each of the above positions. 20. NYIC has also made or anticipates making considerable and additional expenditures for communications, training, and travel expenses to educate members of the immigrant communities we serve about the Census and the citizenship question in particular, including public service announcements, workshops and conferences, and other outreach. We anticipate that these expenditures will total approximately $100,000. 21. NYIC has diverted and anticipates continuing to have to divert a large amount of time from the organization’s managers—including myself, Vice President of Policy Betsy Plum, our communications staff, and managers of member engagement for Long Island and Western New York—to address ongoing concerns related to the citizenship question. 22. NYIC’s increased investment in Census education and outreach work has been driven in

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 280439

part by the concerns of our member organizations and the communities they serve have raised about the citizenship question. NYIC management, including Ms. Plum, remains in regular communication with staff and management at member organizations about the issues and policies affecting immigrant communities in New York. NYIC member organizations, including Chinese- American Planning Council (“CPC”), Arab- American Association of New York (“AAANY”), Masa, Chinese Progressive Association, MinKwon Center for Community Action, and Chhaya Community Development Corporation (“Chhaya”), have reported to Ms. Plum members of the immigrant communities of color they serve expressing an unwillingness to participate in the Census as a result of the citizenship question. In particular, members of immigrant communities of color have expressed significant fear that answering the citizenship question will give a hostile administration information about the number of citizens and noncitizens on a neighborhood basis, or even a city block basis. They are concerned that public citizenship information may be viewed and potentially used for law enforcement profiling against people on blocks with high levels of noncitizens by agencies like Immigration and Customs Enforce- ment or other parts of the Trump Administration that have been used to intimidate and marginalize immigrants. 23. As a statewide organization, NYIC and its members serve a community of approximately four million immigrants across New York. If the addition of a citizenship question is permitted and diminishes the completeness and accuracy of the

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 281440

Census, NYIC, its members, and the immigrant communities we serve will suffer substantial losses of federal resources supporting vital social service, health, education, and other programs supported by Census-guided funds. Moreover, the immigrant communities of color that NYIC and its members serve will also suffer significant diminution of their political power. I declare under penalty of perjury that the foregoing is true and correct. Dated: New York, New York October 26, 2018

/s/ Steven K. Choi

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 282441

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

Civil Action No. 1:18-cv-02921-JMF ______

STATE OF NEW YORK, et al., Plaintiffs.

NEW YORK IMMIGRATION COALITION, et al., Consolidated Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE, et al., Defendants. ______

SUPPLEMENTAL DECLARATION OF STEVEN K. CHOI

I, Steven K. Choi, pursuant to the provisions of 28 U.S.C. § 1746, declare under penalty of perjury that the foregoing is true and correct: 1. As explained in my October 26, 2018 Declaration, I am the Executive Director of the New York Immigration Coalition (“NYIC”). In that capacity, I am responsible in part for NYIC’s

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 283442

education and outreach efforts around the 2020 Decennial Census. I am also one of the NYIC executives responsible for the organization’s budgeting, fundraising, and policy priorities. I have been Executive Director of NYIC for over five years. 2. All of the statements made in my October 26 Declaration and in this Declaration are made based on my personal knowledge, acquired after more than five years as Executive Director of NYIC. During that time period, I have familiarized myself with NYIC’s internal records and processes; our staff and their responsibilities; our programs and program areas; our member organizations including their missions, the communities they serve, and their participation in activities with NYIC. 3. As Executive Director of NYIC, I have spent significant time traveling around New York State and meeting with both our member organizations and the immigrant communities that we serve. Through this process, I have familiarized myself with prevailing views in the community concerning a number of issues, including the Trump Admini- stration’s treatment of immigrant communities and immigration-related policies, and fears in immigrant community about the policies of this Administration. As part of this process, I have gained personal knowledge about the prevailing feeling of fear among immigrant communities in New York State concerning the Trump Admini- stration and, specifically, the decision to add a citizenship question to the 2020 Decennial Census and potential use of the information gleaned from this question. I have also been working with immigrant communities in New York in one form or

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 284443

another since 1999 and have developed a thorough understanding of prevailing views in the communities I’ve worked with. 4. My statements concerning the fears of members of New York’s immigrant community concerning the citizenship question and its effect on their willingness to answer the Decennial Census therefore reflect my knowledge as both the leader of NYIC and a leader in New York’s immigrant community, rather than a restatement of any particular individual’s views. Similarly, in expressing my views about the likely effect of the citizenship question on New York’s immigrant community, I do not intend to offer any specific predictions about non-response rate or percentage undercount, but merely my observations based on my knowledge as a community leader and my work as NYIC Executive Director. 5. I have also been extensively and personally involved in the research, development, and implementation of NYIC’s Census education and outreach programs, including the organization’s work in the New York Counts 2020 coalition. Through that work as well as my work in previous Census cycles, I have gained familiarity with the importance of data gathered from the Decennial Census in apportioning political representation and the allocation of some government funding sources. Because NYIC is driven to address the concerns of its members and the communities they serve in the work that we do, I have also familiarized myself with how those organizations and communities will be impacted by an undercount. Issues related to the Census have been an important part of my work

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 285444

with immigrant communities for three Decennial Census cycles now. 6. My observations in Paragraph 6 of the October 26 Declaration about the Census-guided funding programs in which NYIC member organizations participate is based on my work in the research, development, and implementation of NYIC’s Census education and outreach programs, including the organization’s work in the New York Counts 2020 coalition, as well as my regular and direct involvement with our member organizations, and in particular our Board of Directors and Immigrant Leaders Council. 7. Regarding statements made in Paragraph 7 of my October 26 Declaration, my knowledge that data from the Decennial Census provides the basis for apportioning political representation at the federal, state, and local levels is based on my work in the research, development, and implementation of NYIC’s Census education and outreach programs, including the organization’s work in the New York Counts 2020 coalition, which are high priority programs for NYIC. In addition, my knowledge that an undercount for immigrant communities of color under the Decennial Census would unduly harm our members’ level of political representation at all levels of government is also based on my participation in and management of NYIC’s Census- related work, my past Census work, as well as my familiarity with our member organizations and the communities we serve. 8. Regarding statements made in Paragraph 9 of my October 26 Declaration, I am familiar with NYIC’s activities during the 2010 Decennial

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 286445

Census cycle through my work in the process of researching, developing, and implementing NYIC’s education and outreach for the 2020 Decennial Census cycle and also my work on the 2010 Decennial Census cycle for one of NYIC’s member organizations, MinKwon Center for Community Action. 9. Regarding statements made in paragraph 14 of my October 26 Declaration, my observation that New York immigrant communities’ heightened fear of interacting with government workers has increased even further due to the decision to add the citizenship question is based on knowledge that I have gained in the course of developing, implementing, and supervising NYIC’s Census education and outreach work with its members and the communities they serve, including NYIC’s work with the New York Counts 2020 coalition, as well as my extensive contact with immigrant communities throughout New York as Executive Director of NYIC and as a leader in New York’s immigrant community. This body of knowledge, as well as my engagement with immigration policy as Executive Director of NYIC, also supports my statements about the intrusiveness of the citizenship question into immigrant communities and the hostility of the Trump Administration towards those communities. 10. Regarding statements made in paragraph 15 of my October 26 Declaration, my observation that NYIC has faced, and will continue to face, a more difficult Census-response environment due to New York immigrant communities’ heightened fear of interacting with government workers because of

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 287446

the addition of the citizenship question is based on based on my research of NYIC’s past Census efforts for developing and implementing NYIC’s 2020 Census activities, my work with MinKwon during the 2010 Decennial Census cycle, my experience in developing NYIC’s Census work before the decision to add a citizenship question, as well as my participation and supervision in NYIC’s extensive Census work since the citizenship question was announced. Census education and outreach programs for immigrant communities for the 2020 Decennial Census have been on NYIC’s policy agenda since at least the middle of 2017. 11. Regarding statements made in paragraph 22 of my October 26 Declaration, my statements about information I have received from NYIC staff, including Vice-President of Policy Betsy Plum, regarding the concerns of NYIC member organizations and the communities we serve illustrate part of the basis for my decisions as Executive Director in authorizing NYIC’s increased investment in Census education and outreach work. These decisions were also informed by my own work and observations in working with NYIC staff and member organizations-in particular, our Board of Directors and Immigrant Leaders Council—in researching, developing, and implementing our education and outreach programs for the 2020 Decennial Census cycle, as well as my interactions with immigrant communities in my role as a community leader. This body of knowledge also informs my understanding that immigrant communities are color are reluctant to participate in the Census and are concerned about the intrusion into their privacy that may result from

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 288447

the publication of data derived from the citizenship question. 12. Regarding statements made in paragraph 23 of my October 26 Declaration, my statements that about the effect of any undercount of immigrant communities is based on my participation and supervision of the research, development, and implementation of NYIC’s Census education and outreach programs, through which I have gained an understanding of the role of Census data m the apportionment of political representation and the allocation of government funding. Again, in expressing my views about the likely effect of the citizenship question on New York’s immigrant community, I do not intend to offer any specific predictions about non-response rate or percentage undercount, but only my observations based on my knowledge as a community leader and my work as NYIC Executive Director, and my participation and supervision of our extensive education and outreach efforts for the 2020 Decennial Census. I declare under penalty of perjury that the foregoing is true and correct. Executed: November 2, 2018 New York, NY /s/ Steven K. Choi

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 289448

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

Civil Action No. 18-cv-2921-JMF Hon. Jesse M. Furman ______

NEW YORK IMMIGRATION COALITION, et al., Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE, et al., Defendant. ______

DECLARATION OF JENNIFER VAN HOOK

1. I, Jennifer L. Van Hook, pursuant to the provisions of 28 U.S.C. § 1746, declare under penalty of perjury that the following is true and correct. 2. I submit this declaration in lieu of direct expert testimony in the trial in the above captioned cased.

I. Background and Qualifications 3. I was asked by Plaintiffs to bring my scientific expertise and experience to bear on the

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 290449

question of whether the addition of the citizenship question to the 2020 Decennial Census is likely to result in greater item and unit nonresponse, particularly among populations for whom the citizenship question may be particularly sensitive. 4. I am Roy C. Buck Professor of Sociology and Demography at the Pennsylvania State University. I served as director of the Population Research Institute at Penn State from 2011 through 2016. Currently, I am the director of graduate studies in sociology at Penn State and co-editor of Demography, the flagship journal for population science. I am also a non-resident fellow at the Migration Policy Institute. 5. I am trained as a sociologist and demographer. I obtained a PhD in Sociology in 1996 from the University of Texas at Austin. I have an M.S. in Sociology from the University of Wisconsin at Madison and B.A. from Carleton College. After obtaining my PhD, I worked at the Urban Institute on projects related to education and program participation among immigrants. In 1999 I joined the faculty at Bowling Green State University, and then moved to Penn State University in 2007. 6. I have over 20 years of research experience analyzing large demographic data sources on topics related to immigration. My publications have appeared in major sociology and demography journals, including Demography, Journal of Health and Social Behavior, Social Science and Medicine, Sociology of Education, Social Forces, and American Sociological Review, and I have received external funding for my work from the National Institutes of Health, the National Science Foundation, the

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 291450

Foundation for Child Development, the Russell Sage Foundation, and the U.S. Census Bureau. 7. My work uses demographic methods to estimate the size, characteristics, and dynamics of the foreign-born population. My colleagues and I have evaluated and improved estimates of the unauthorized foreign-born population. This line of research resulted in several high-profile publications, including new estimates of the size and heterogeneity of the unauthorized Mexican-born population (Bean et al. 2001); the development of a new method and estimates of foreign-born emigration (Van Hook et al. 2006; Van Hook & Zhang 2011) and coverage error (Van Hook et al. 2014); new assessments of the quality of self- reported data on citizenship and legal status (Van Hook and Bachmeier 2013; Bachmeier, Van Hook and Bean 2014); and monte carlo simulations that tested a variety of legal status imputation approaches (Van Hook et al. 2015). The work on legal status led to important innovations that have enabled researchers at the Migration Policy Institute and elsewhere to produce estimates of the characteristics and geographic distribution of the unauthorized population in greater detail than possible with earlier methods. A copy of my Curriculum Vitae includes a complete list of my publications. See PX-536. 8. I served as a member of the Census Advisory Committee of Professional Organizations, PAA, from 2008 to 2011. I also served as an expert for the 2010 Census Demographic Analysis Program (Net International Migration Team) and am currently serving on the 2020 Census Demographic

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 292451

Analysis Program (Net International Migration Team). In such capacities, I advise the Census Bureau on various issues, including which items should be included in the decennial Census and the likely impact of adding such questions on the overall quality of the data collected. 9. I submitted an expert report in this case on September 7, 2018. See PX-317. 10. I submitted a supplemental expert report on October 23, 2018, based on data that was not available when I drafted my initial report, including American Community Survey data made available to the public by the Census Bureau just recently on October 18, 2018. See PX-318. 11. Based on my experience, training, knowledge, and education, I believe I am well qualified to offer expert opinions on the question of whether the addition of the citizenship question to the 2020 Decennial Census is likely to result in greater item and unit nonresponse, particularly among populations for whom the citizenship question may be particularly sensitive. I hold my opinions in this case to a strong degree of professional certainty.

II. Summary of Opinions 12. My analyses of the patterns and trends in nonresponse point to three key findings: 1) Hispanics and immigrants (especially noncitizens or those living in immigrant households) tend to have relatively high unit nonresponse rates and item nonresponse rates for questions on place of birth and citizenship.

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 293452

2) Both unit and item nonresponse have increased significantly over time among immigrants. Unit nonresponse rates rose in the Trump era among noncitizens while remaining steady or declining among citizens. Item nonresponse also rose among those living in immigrant households, particularly for Hispanics. 3) Hispanic immigrants experienced particularly large increases in nonresponse since the start of 2018. 13. Between the last quarter of 2016 and the first quarter of 2018, the adjusted unit nonresponse rate increased among Hispanic noncitizens by 43 percent. Adjusted item nonresponse rates increased by a similar amount — 45 percent — among Hispanics in immigrant households during the same time period. These are large changes for such a short period of time, especially when considering that my analysis probably underestimates the true levels of unit nonresponse in the population. As I discuss in greater detail below, the outcomes I examined — unit nonresponse for follow-up interviews and item nonresponse — are measured on a sample of people who have already agreed to participate in at least one Current Population Survey (“CPS”) interview. My results do not capture the patterns and trends in nonresponse for the least compliant survey respondents, those who never agreed to participate in the CPS. 14. While I cannot state with certainty which factors or events contributed to these trends, my analyses controlled for and enabled me to discount several of the most mundane explanations,

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 294453

including the possibilities that they arose from changing demographic composition of the population, household migration patterns, sudden up- ticks in household migration or deportations, or that they reflect increasing trends in item nonresponse in general and not just for citizenship/ nativity questions. 15. Instead, the results of my analyses are consistent with the concept that immigrants, especially Hispanic noncitizens and those living in foreign-born households, have become less compliant respondents to Census Bureau surveys since early 2017, and more sensitive in particular to question concerning citizenship. Moreover, the timing of the sharp increase in item and unit nonresponse in 2018 is noteworthy as it overlaps with the time period when the 2020 citizenship question was being discussed in national media and when the issue of citizenship appeared to be particularly salient among Spanish-speaking internet users. 16. In sum, the overall patterns of unit and item nonresponse are consistent with the understanding that survey response changed for the worse among Hispanic noncitizens, and the timing of the change suggests that it is linked to changes in the political and/or policy climate, including the proposal to add the citizenship question to the 2020 Decennial Census. The unit nonresponse data is consistent with the understanding that adding a citizenship question to the Census will reduce nonresponse rates among racial minorities (in particular Hispanics) as compared to non-Hispanic whites, and noncitizens as compared to citizens. Moreover, the item non-response rate estimates raise

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 295454

concerns that adding a citizenship question will not yield accurate citizenship responses for a substantial percentage of certain groups, including Hispanics, Asians, and people living in a household with at least one immigrant.

III. Overview of Analysis 17. For my initial expert report, I analyzed data from the Current Population Survey (“CPS”), which is a monthly survey of approximately 60,000 U.S. households administered by *** 66. Analysis of the second quarter 2018 CPS data confirmed trend observations and opinions in my initial report. As previously discussed, item nonresponse to citizenship changed slowly until about 2016, after which it increased rapidly for Hispanics, particularly those in immigrant households, reaching its peak in the first two quarters of 2018, while there was no corresponding increase in item nonresponse on questions about age. See Table A1-S and Figures 13-S and 14-S in the Appendix. 67. These patterns remained even after adjustments were made to account for shifts in population composition with respect to state of residence, age, sex, education and nonresponse on age, sex, and month of interview. See Table A2-S and Figures A2-S and A3-S in the Appendix. During the first half of 2018, the CPS adjusted item nonresponse rates to citizenship questions of Hispanics (7.9 percent) and of Hispanics in immigrant households (10.5 percent) were significantly greater than the respective rates in all of the years from 2013 to 2016.

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 296455

68. I also conducted a difference-in-change analysis of CPS item nonresponse using the adjusted data and found that the change in item nonresponse for Hispanics between 2016 and June 2018 is significantly greater than the change from 2013 to 2015. This is shown in Table S1 in the Appendix. This acceleration of the item nonresponse rate can also be seen among Hispanics living in immigrant households, leading this group to have significantly higher item nonresponse in 2017 and 2018 than in any year from 2013 to 2016 (see Table A2-S and Figure 14-S in the Appendix).

B. ACS Item Nonresponse 69. The American Community Survey (ACS) is important because it is the most similar to the 2020 Census with respect to its mode of data collection. Also, similar to the decennial census, response to the ACS is mandatory by law. Dr. Abowd’s January 19, 2018 memorandum to Secretary Ross analyzes ACS data for 2013-2016. I conducted an analysis based the more current 2017 ACS data, which only recently was made available to the public by the Census Bureau on October 18, 2018. 70. Item nonresponse on citizenship increased significantly among Hispanics each year from 2013- 2017 as shown in Figure S1 below (and Table S3 in the Appendix). In contrast, item nonresponse did not significantly increase after 2014 among Asians, and it did not significantly increase after 2015 among blacks and among non-Hispanic whites. By 2017, item nonresponse on citizenship was significantly higher among Hispanics than Asians, blacks, and non-Hispanic whites.

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 297456 2017 Year, ■ 4.7 4.6 and 4.8 2016 site) ■ NH-Whites 4.3 2015 ■ download FTP s Race/Ethnicity 2014 ■ by Censu the ACS 2013 Blacks weights) 18+ from the in Age replicate ACS 7.8 (downloaded Adults 6 . 7 using files Citizenship 7.5 ACS on Among Asians I (calculated I 6.8 microdata vals use - inter T 8.2 Non-Response public 1% Item confidence % 2017 - Hispanics 95 S1. 2013 I 5.8 bars= Figure 2 1 9 3 0 7 6 5 8 4 Error Source: 10

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 298457

71. Among those in immigrant households, blacks have higher item nonresponse than Hispanics as shown in Figure S2 below (and Table S3 in the Appendix). This differs from all adults, among whom Hispanics have the highest nonresponse rates. However, the temporal patterns among those living in immigrant households are similar as for all adults. Item nonresponse increased significantly every year from 2013 to 2017 among Hispanics as shown in Figure S2. In contrast, item nonresponse did not significantly increase after 2014 among Asians; it did not significantly increase after 2015 among blacks; and it did not significantly increase in any year among non-Hispanic whites.

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 299458 Year, and site) NH-Whites I 6.6 download FTP l 10.5 Race/Ethnicity l 3 . by 10 _1_ I Census I households L the 10.2 _ I ACS : Blacks weights) from 9.4 the in I I immigrant replicate in 8.0 ACS (downloaded I I 18+ 7.8 files using Citizenship II 7 Age . 2017 ACS on I ■ ! Asians 6 7 . - 7 I (calculated Adults 2016 I microdata ■ I use - intervals Among 2015 9.3 i I Non-Response ■ I public 8.5 r¼-, 1% I~ Item 2014 - confidence 7.5 - 017 ■ % -2 Hispanics I S2. 95 = 2013 2013 I : 6.6 bars Figure 2 0 6 4 8 Error Source 12 10

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 300459

72. I also examined how item nonresponse on citizenship compares with the other decennial census questions that are asked in the ACS: age, sex, race, Hispanic origin, and housing tenure (rent/own). In the 2017 ACS, for all four racial/ ethnic groups, item nonresponse rates on citizenship was significantly higher than any other decennial census item as shown in Figure S3 and Table S4 in the Appendix. 73. Additionally, I examined trends in item nonresponse on all decennial census questions that also are on the ACS among Hispanics to assess whether the increase in nonresponse on the citizenship question is part of a broader pattern of increasing nonresponse. My analysis shows that this question is different from the other decennial census questions. The percentage point change in item nonresponse on the citizenship question increased since 2013 more than it did for the other questions, as shown in Figure S4 below (and Table S5 in the Appendix).

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 301460 Sex Age Race Origin Hispanic Tenure Items, Citizenship Census 2017 Decennial ACS on ite) s 2016 2013-2017 download 18+, FTP Non-Response s u Age s Cen Item the in Adults 2015 weights) from 2013 replicate Hispanic Since ACS s (downloaded ing s file u Among Change ACS 2014 Year, Point (calculated by microdata use - intervals public Percentage % 1 2013 confidence S4. % 2017 - 95 = s 2013 Figure : 0 5 bar . . 2 2.0 1.5 1.0 3.0 0.5 0 -0.5 -1.0 Error Source

78228 • ACLU • APPENDIX 8 (DOC 53-55) AL 3/24/19 302461

74. Finally, I conducted a difference-in-change analysis of ACS item nonresponse on citizenship shown in Table S6 in the Appendix. Item nonresponse increased more for Hispanics as a whole as well as for Hispanics in immigrant household between 2015 and 2017 than in the earlier 2013 to 2015 time period. 75. In sum, the item nonresponse trends I identified in my analysis of the ACS data are consistent with the findings of my analysis of the CPS data.

VII. Conclusion 76. In sum, it is my opinion to a strong degree of professional certainty that the overall patterns of unit and item nonresponse seen recently in CPS and ACS surveys are consistent with the understanding that response to Census Bureau surveys has changed for the worse among Hispanics, particularly those in immigrant households. The timing of the change suggests that it is linked to changes in the political and/or policy climate, including the proposal to add the citizenship question to the 2020 Decennial Census. The unit nonresponse data is consistent with the understanding that adding a citizenship question to the Census will reduce nonresponse rates among racial minorities (in particular Hispanics) as compared to non-Hispanic whites, and among noncitizens as compared to citizens. Moreover, the item nonresponse rate estimates raise concerns that adding a citizenship question will not yield accurate citizenship responses for a substantial percentage of certain groups, including Hispanics, Asians, and people living in a household with at least one immigrant.

78228 • ACLU • APPENDIX 8 (DOC 53-55 xyz 00:00 00/00/07 303462

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

CIVIL ACTION NO. 1:18-cv-2921 (JMF) ______

STATE OF NEW YORK, et al., Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE; et al., Defendants. ______

AFFIDAVIT OF SUSAN BROWER

Pursuant to 28 U.S.C. § 1746(2), I, SUSAN BROWER, hereby declare as follows: 1. I am over the age of eighteen and have personal knowledge of all the facts stated herein. 2. I obtained my Ph.D. in sociology from the University of Michigan. 3. I am the State Demographer for the State of Minnesota. As State Demographer, I oversee the Minnesota State Demographic Center (“the Center”).

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 304463

By statute, I am appointed by the Commissioner of Administration. Minn. Stat. § 4A.02(a). My duties are identified in Section 4A.02(b), and include gathering and developing demographic data relevant to the state; serving as the liaison to the United States Bureau of the Census; coordinating state and federal demographic activities; and aiding the state legislature in preparing a census data plan and form for each decennial census. I have been employed as the Minnesota State Demographer since 2012. My knowledge of the facts contained herein is based on my experience as a professional demographer, and in my capacity as the Minnesota State Demographer. My knowledge is also based on records I reviewed in my capacity as State Demographer, including records on Minnesota’s redistricting, and on publically available data regarding the population and demographics of the State of Minnesota. 4. As liaison between the Census Bureau and the State of Minnesota with respect to redistricting, I inform the Census Bureau about the type of and format of data the State of Minnesota needs and wants to complete its redistricting. In order to identify what the State needs and wants in connection with redistricting, I anticipate commu- nicating with the State of Minnesota legislative redistricting committees. Those committees have not been formed yet for the 2020 Census; based on the process for redistricting following the 2010 census, I expect each party caucus (one Republican and one Democratic Farmer Labor (“DFL”)) to form one committee per legislative chamber, such that there are four redistricting committees. I also anticipate working with the redistricting committees

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 305464

to send them the census data file from the Census Bureau and help them understand the data. 5. The Center works to ensure a full count and reduce non-response rates and undercounting. The Center’s outreach efforts are premised in part on providing residents accurate answers to their questions about the census. Residents have already asked questions about the impact of the citizenship question, and I expect those questions to continue. One question residents have asked relates to the confidentiality of survey responses. Census surveys are supposed to be confidential. That is one mecha- nism to encourage full participation to obtain an accurate count. 6. However, even if a specific person’s survey answers will not be public, some census data is generally made public on a very granular level. The U.S. Census produces some data at the block level. The block level is a low level of geography. For example, according to the 2010 Census Redistrict- ing Data (Public Law 94-171) File, 26,213 blocks in Minnesota contained only one household; 95,220 blocks in Minnesota contained between one and nine households. That compares to the total of 151,646 blocks in Minnesota that contained more than zero households. In other words, 63% of the blocks in Minnesota that contained any households, contained fewer than ten households. 7. Relatedly, 5,685 blocks in Minnesota contained only one person; 87,135 blocks in Minnesota contained between one and nineteen people. That compares to 151,983 blocks in Minnesota that contained more than zero persons. In other words, 57% of the blocks in Minnesota that contained any

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 306465

people contained fewer than twenty people. If the citizenship question is asked, therefore, the public would be aware on a block level, how many residents are non-citizens . 8. At a meeting with U.S. Census Bureau officials the week of July 30, 2018, in Los Angeles, I conveyed that Minnesota does not want to receive citizenship status at the block level. I did so because it is not necessary for redistricting, I believe that publishing that data could undermine the confidential nature of the survey, and I believe aware- ness that such data will be published could discourage residents from completing their census surveys. 9. I was informed by U.S. Census staff that even if the data was not published to the states in the form of a data file in connection with redistricting, if the Census gathers citizenship data, then it will make that data public on a block level. If I know that citizenship data will be published at a block level, it will be harder to give people assurances that their survey responses will be safe and confidential. Simply, the disclosure of block-level citizenship data will adversely affect our outreach and likely our response rates. I declare under penalty of perjury that, to the best of my knowledge, the foregoing is true and correct.

Executed on this 25 day of 10, 2018

/s/ SUSAN BROWER

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 307466

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

CIVIL ACTION NO. 1:18-cv-2921 (JMF) ______

STATE OF NEW YORK, et al., Plaintiffs, New York Immigration Coalition, et al., Consolidated Plaintiffs —v.—

UNITED STATES DEPARTMENT OF COMMERCE; et al., Defendants. ______

DECLARATION OF GEORGE ESCOBAR

I, George Escobar, pursuant to the provisions of 28 U.S.C. § 1746, declare under penalty of perjury that the foregoing is true and correct: 1. I am the Chief of Programs and Services at CASA (“CASA”). In this capacity, I oversee CASA’s services departments, including legal services, health services, workforce develop-

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 308467

ment, employment, education, and our immigrant integration programs. 2. Based on my nearly 20 years of professional experience working for Latino and immigrant serving organizations in both the public and private sectors, the current outreach to immigrant communities I oversee in my current capacity at CASA, and my conversations with individual CASA members, I believe the addition of a citizenship question to the 2020 Decennial Census short-form questionnaire will deter participation of many individuals in the Latino and immigrant communities and result in a disproportionate undercount of Latinos. As a result of the decision to add this question, CASA has diverted and will continue to be required to divert resources from core organizational priorities to additional, remedial outreach to counteract the citizenship question’s negative effect on Census response rates in our community. 3. This undercount will also injure many CASA members, many of whom live in areas of the United States in which immigrants of color and other communities of color exceed national and state averages. Specifically, they will be harmed because the disproportionate undercount that will result from the citizenship question among immigrants of color will diminish their political power relative to other parts of their states. In addition, this undercount will diminish the amount of Census- related funding those areas receive for key programs, services, and facilities that our

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 309468

members and their families rely on to succeed in the communities in which they live.

I. CASA’s Mission and Activities 4. CASA in a non-profit 50l(c)(3) membership organization headquartered in Langley Park, Maryland, with offices in Maryland, Virginia, and Pennsylvania. Founded in 1985, CASA is the largest membership-based immigrants’ rights organization in the mid-Atlantic region, with more than 90,000 members. 5. CASA’s mission is to create a more just society by increasing the power of and improving the quality of life of low-income immigrant communities. To advance this mission, CASA offers social, health, education, job training, employment, and legal services to immigrant communities. CASA serves nearly 20,000 people a year through its offices and provides support to additional clients over the phone and through email. 6. As Chief of Programs and Services, my fundamental role in the organization is to align its programming with its mission and ensure our interventions and programs are effective in addressing disparities impacting immigrant communities while setting our members and their families on a path to success.

II. CASA’s Census-Related Work 7. CASA has an ongoing commitment to promoting engagement in the Decennial Census among its members, constituents, and communities. Member participation in the Decennial

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 310469

Census advances CASA’s mission by increasing the political power of low-income immigrant communities and improving quality of life for those communities through increased population- driven government funding. 8. Given CASA’s 30 year history working with immigrant communities throughout the Mid- Atlantic region and the trust it has developed with these communities as a result of this history, CASA has consistently been a “go-to” organizational partner in conducting outreach and education around the Decennial Census and other related activities. 9. In 2010, CASA partnered with various local government education and outreach campaigns throughout Maryland. In addition, CASA received dedicated funding from other sources to conduct door-to-door outreach, facilitate group educational sessions, and work with local, ethnic media to inform, engage, and encourage participation in the Census among Limited English Proficient, immigrant communities in the region.

III. CASA’s Work Connecting Members with Government Services 10. CASA provides a number of services screening, navigating and enrolling its members for public benefits for which they are eligible. This work includes connecting members with public benefits related to health, education, and legal services. 11. Through its Health and Social Service Program, CASA case managers assess its members needs

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 311470

and screen for eligibility for several public benefit programs. CASA assists eligible members in enrolling in food assistance programs such as the Supplemental Nutritional Assis- tance Program (SNAP), which provides food assistance to low-income individuals, and the Special Supplemental Nutrition Program for Women Infants and Children (WIC), which provides supplemental foods, health care referrals, and nutrition education to eligible individuals. 12. CASA also assists eligible members in enrolling for health care coverage through Medicaid, or enrolling their children in the Children’s Health Insurance Program. 13. CASA’s Education Department works with families residing in dense immigrant neighborhoods to navigate them through the school enrollment process in Title 1 funded public schools serving these neighborhoods. Staff also provide information about early childhood education programs available in the neighborhoods served by CASA and partner with local providers to enroll eligible children in those programs. 14. In addition, CASA’s Legal Program provides individualized legal counseling to individuals claiming employment discrimination, wage theft, and other issues that may lead to job loss. For those who have suffered from job loss, CASA provides assistance in navigating the unemployment insurance application process for those who are eligible.

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 312471

IV. Harm to CASA as a Result of the Citizen- ship Question 15. As described below, the addition of a citizenship question to the Decennial Census harms CASA members, as well as the organization itself. The increased fear among immigrants, caused by the Trump Administration’s anti- immigrant policies and rhetoric, has led to decreased engagement with the government generally, and specific fear of participating in the Decennial Census. This fear and decreased participation harms CASA as an organization, because we are now diverting our limited resources in an effort to encourage participation in the Decennial Census.

A. Fear in the Latino Immigrant Community under the Current Administration 16. Based on my conversations and work in the community, immigrant members of CASA have felt targeted, bullied and attacked by the rhetoric and actions of the Trump Administra- tion. Starting with the presidential campaign, through to its current policies, CASA members have expressed fear, frustration and despair over how they feel this administration has scapegoated them. In particular, they have expressed fear over the emphasis this administration has placed on increased deportations and immigration enforcement. 17. In particular, policies such as family separation at the border, the “zero-tolerance policy” for immigrants entering the United States

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 313472

without documentation, and the elimination of the Deferred Action for Childhood Arrivals (DACA) and Temporary Protected Status (TPS) programs have all increased the fear among immigrant communities of being targeted by agencies such as Immigration and Customs Enforcement (ICE). This fear has led to a chilling effect on the willingness of immigrant communities of all types, regardless of immigration status, to interact with the federal government in any way. 18. The recent announcement of changes to the definition of who constitutes a “Public Charge” for immigration purposes has only inflamed the current situation. Collectively, these policies have led to widespread confusion and fear, to the point that many individuals are hesitant to apply for a public benefit for which they are eligible to receive. 19. CASA and its partner providers have begun to track a decrease in the number of individuals applying for certain public benefits. Most troubling has been a decrease in the number of enrollments into benefits completely unrelated to the policies that have been announced, such as a decrease in the number of immigrants applying for health insurance made available through the Affordable Care Act or a decrease in the number of Legal Permanent Residents applying for citizenship. To those of us at CASA, this decreased participation indicates that the Trump Administration’s policies are not only affecting those individuals and communities who are directly targeted by these

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 314473

policies, but are increasing fear among the immigrant community more broadly.

B. CASA Members are Less Willing to Engage with the Decennial Census due to Increased Fear 20. Under the current environment, CASA members have expressed fear of even a knock on their door by a stranger given the high number of arbitrary immigration enforcement actions that have resulted in the deportation of immigrants with little or no criminal background in the communities CASA serves. This fear of even speaking to someone approaching their door will certainly be compounded by a request by a government official, such as a census enumerator, regarding an individual’s citizenship status, or the citizenship status of others in their household. 21. Many CASA members have expressed doubts and fears to me about how information within the government is shared, and whether an answer to a question on a particular document or application may lead to their family being harmed or separated. Under these circum- stances, many have expressed a feeling that participation in the Census presents too high a risk to the safety and security of themselves and their families to justify participating. 22. The fear of participating in the Census has been shared with multiple CASA staff and volunteers who routinely conduct field outreach in the community, such as CASA Health Promoters and Community Organizers. Shortly

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 315474

after information about the addition of the citizenship question became public and was reported on by Spanish language media, these outreach workers reported that the reaction of fear in the community was widespread. Many CASA staff members and volunteers reported an increase in general suspicion and fear among community members in completing any type of application or document associated with any government entity. 23. Recently, a CASA Health Promoter shared an interaction with me that she had with a CASA member who told her a story about a family member that he believed was placed in deportation proceedings shortly after applying for a Driver’s License. “If those two things are linked,” the member reportedly said, “then why on earth would I answer a question about immigration status on a form from the federal government? It’s not worth the risk.” That reaction is typical of what we have experienced in the community. 24. Through this litigation, I have learned that block-level data can identify the characteristics of a population within a very small geographic area. By identifying the citizenship status of individuals in such a small area, I am concerned that this could lead to immigration enforcement targeting these particular areas and communities, thus harming the privacy of CASA members. I believe this concern is particular will discourage CASA members from participating in the Decennial Census.

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 316475

C. CASA is Diverting Resources to Encourage Participation in the Decennial Census 25. CASA recognizes the importance of the Decennial Census for its members and the broader Latino immigrant community. Parti- cipation in the Census is necessary to preserve the political power of these communities, and to ensure that these communities receive the vital federal resources that so many of our members rely upon. In order to combat the increased fear in the Latino immigrant community, and the unwillingness of many CASA members to participate in the Decennial Census, CASA is currently planning a massive response to try and overcome the many barriers to participation. 26. In order to combat these barriers, CASA will have to reorganize its communication team and reassign staff to Census outreach and education to a level not previously anticipated. Given significant funding for this work is not anticipated, much of the work will have to be funded through other sources, perhaps including CASA reserves and leveraged with volunteers. 27. Our work in response to the addition of the citizenship question on the Decennial Census has already begun. CASA’s Community Organizing team has already conducted several “house” meetings on this topic to combat the fear and suspicion that has already been expressed about the addition to the citizenship question. These meetings have been held in many of the neighborhoods regularly canvassed

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 317476

by the community organizing team and have been focused on the importance of participating in the census. 28. Education about the census is also being imbedded within the “Get Out The Vote” canvassing work being done by CASA during the 2018 election season. CASA canvassers are provided talking points and some literature about the Census to share with voters as they conduct outreach in their work encouraging voters to cast their ballots. 29. The increased focus on Census outreach and education will necessarily divert CASA’s limited resources, resulting in less resources being allocated to other vital CASA programs. 30. I declare under penalty of pc1jury that the foregoing is true and correct.

Executed: October 26, 2018 Washington, DC

/s/ George Escobar

78228 • ACLU • APPENDIX 09 (DOCs 58-63) AL 3/24/19 318477

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

CIVIL ACTION NO. 1:18-cv-2921 (JMF) ______

STATE OF NEW YORK, et al., Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE; et al., Defendants. ______

AFFIDAVIT OF MARCHELLE FRANKLIN

Pursuant to 28 U.S.C. § 1746(2), I, Marchelle Franklin, hereby declare as follows: 1. I am over the age of eighteen and have personal knowledge of all the facts stated herein. 2. I am the Human Services Director for the City of Phoenix, a role I have held since November 2017. In this role, I oversee a staff of 375 employees and a budget of approximately $84 million. Part of my work includes overseeing

78228 • ACLU • APPENDIX 10 (DOC 60) AL 3/24/19 319478

the federally-funded programs described in this declaration. 3. I have worked for the City of Phoenix since 2002. My previous roles include serving as the Interim Human Services Director, Director of the Police Department’s Community Affairs Division, and Chief of Staff for former Mayor Phil Gordon. I have a Masters of Business Administration in finance from Grand Canyon University, a Bachelor of Science in business administration from the University of Phoenix, and I completed the Executive Development Program in human resources consulting and strategy development at the University of Southern California. 4. The City of Phoenix Human Services Depart- ment provides a comprehensive array of services to help people meet emergency, short- and long-term needs, and help every individual reach their highest level of self-sufficiency. Program areas include early childhood education, emergency assistance, older adult services, crime victim services, homeless services, business/workforce development and community initiatives: a. Education. Responsible for overall implementation and monitoring of Head Start performance standards for more than 3,300 children through a directly operated program and contracted delegate agencies. b. Community and Senior Services. Community Services develops, implements and operates human service programs to meet the emergency, short and long-term

78228 • ACLU • APPENDIX 10 (DOC 60) AL 3/24/19 320479

needs of eligible low-income residents, to include the elderly and individuals with disabilities. c. Family Advocacy Center. The Family Advocacy Center uses a multidisciplinary approach to provide comprehensive services to victims of violent crime such as domestic and sexual violence, child abuse, physical assault and homicide. d. Homeless Services. Homeless Programs provide emergency shelter, rapid-rehousing (Housing First), street outreach and supportive services components for homeless families, youth and individuals through city programs and through contracts with local community providers. e. Business and Workforce Development. Business and Workforce Development programs provide job readiness and skills training for adults and youth. 5. An undercount of the true population of the City of Phoenix in the upcoming 2020 Census would impact the work of the Human Services Department in a number of ways. 6. Victims of Crime Act (“VOCA’’) Funds. The U.S. Department of Justice administers the distribution of VOCA funds. The VOCA statutory distribution formula provides each state with a base amount for each state and distributes the remainder proportionately, based on U.S. Census population. These funds support direct services to victims of crime that assist with emotional, psychological or physical

78228 • ACLU • APPENDIX 10 (DOC 60) AL 3/24/19 321480

needs; stabilization after victimization; understanding and participating in the criminal justice system; and/or restoring a measure of security and safety. A differential undercount on the 2020 Census that impacts Arizona relative to other states would likely affect the VOCA funding passed through to the City of Phoenix that provide crucial services to victims of domestic violence, sexual assault, and physical assault. The services provided include crisis intervention, emergency services (i.e. transportation, child care services, and temporary housing), and comprehensive assistance with navigating the criminal justice system. The City of Phoenix’s Family Advocacy Office receives $ 665,420 annually in VOCA funding. A 1% cut in VOCA funding would make $6,654 less funding available for the provision of these services. 7. Temporary Assistance for Needy Families (“TANF”) Funds. The U.S. Department of Health and Human Services distributes TANF funds to the State of Arizona, who then distributes these funds to municipalities or other organizations based on the number of low-income individuals under 125% of the Federal poverty level in each service area. TANF funds are used to design and operate programs that help needy families achieve self-sufficiency. An undercount of the population in Phoenix would likely affect the funding provided to the City of Phoenix to provide case management services (counseling, skills development, budgeting, employment assistance, etc.) to individuals and families seeking self-

78228 • ACLU • APPENDIX 10 (DOC 60) AL 3/24/19 322481

sufficiency. The City of Phoenix’s Community and Family Services Centers use $985,710 in TANF funds annually. A 1% reduction in those funds would be a loss of $9,857, which would translate into a loss in man hours, negatively impacting the City’s ability to provide critical services to its population. 8. Low-Income Home Energy Assistance (“LIHEAP”) Funds. The U.S. Department of Health and Human Services distributes LIHEAP funds to the State of Arizona, who then distributes these funds to municipalities or organizations based on the number of low- income individuals under 125% of the Federal poverty level in each service area. LIHEAP funds are used to assist families with energy costs. A differential undercount in Arizona relative to other states, or in Phoenix relative to other municipalities in Arizona, would likely affect the funding provided to the City of Phoenix to provide assistance with paying utility bills, increasing the risk of health and safety problems that arise from unsafe heating and cooling practices. The City of Phoenix’s Community and Family Services Centers have an annual LIHEAP budget of $6,020,953. A 1% reduction would result in the loss of $60,209 and would mean utility assistance payments will not be processed for roughly 109 households. 9. Head Start Funds. The U.S. Department of Health and Human Services grants funding directly to municipalities and other organizations to promote school readiness of children

78228 • ACLU • APPENDIX 10 (DOC 60) AL 3/24/19 323482

from birth through age 5 from low-income families through education, health, social and other services. Each municipality/non-profit providing Head Start services is required to use a community needs assessment in order to identify eligible families for selection and enrollment. Head Start slots are placed in specific areas related to where eligible families are in the city. The families we serve are at poverty and have additional risk factors. There is no requirement for them to have legal presence. The community needs assessments require the use of census data to identify where we place our services and slots. A differential undercount of the population of children aged 0-5 in Phoenix, relative to other municipalities, would likely affect the community needs assessment, which will lead to erroneous information that will impact identifying families that meet the eligibility requirements for Head Start and associated services. A 1% cut in the City’s $34,405,759 annual Head Start budget would mean a loss of $344,057 and result in the loss of 35 slots for these much-needed school readiness services. 10. Workforce Innovation and Opportunity Act (“WIOA”) Funds. The U.S. Department of Labor grants funding to the State of Arizona, who then distributes to municipalities and other organizations based on: (a) local area relative share of total unemployed in area of substantial unemployment; (b) local area relative share of excess unemployed; and (c) local area relative share of economically disadvantaged adults ages 22-72. A differential

78228 • ACLU • APPENDIX 10 (DOC 60) AL 3/24/19 324483

undercount of the population of the City of Phoenix relative to other municipalities will impact the distribution of funds to the City of Phoenix by reducing the number of partici- pants who would receive services under the WIOA programs and services such as employment training, skills development, occupational training. Stated differently, because allocations are based on formulas stated above, an undercount of individuals associated with a particular metric/data point will reduce the allocation, thereby reducing the number of individuals that can be served with those program funds. These programs also include a youth component designed to prepare youth to enter post-secondary education, training and employment. The City of Phoenix receives $10,751,819 in WIOA funds annually. A 1% reduction in those funds would result in $107,518 less funds available for employment and support services to adult, youth, and displaced workers. I declare under penalty of perjury that, to the best of my knowledge, the foregoing is true and correct. Executed on this 25th day of October, 2018

/s/ MARCHELLE FRANKLIN

78228 • ACLU • APPENDIX 10 (DOC 60) AL 3/24/19 325484

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

CIVIL ACTION NO. 1:18-cv-2921 (JMF) ______

STATE OF NEW YORK, et al., Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE; et al., Defendants. ______

AFFIDAVIT OF EMILY FREEDMAN

Pursuant to 28 U.S.C. § 1746(2), I, Emily Freedman, hereby declare as follows: 1. My name is Emily Freedman. I am over the age of eighteen and have personal knowledge of all the facts stated herein. 2. I am Director of Community Development for the City of Providence, Rhode Island. I oversee the administration, coordination, and preparation of all work required to complete and fulfill the City’s obligations under its community development and

78228 • ACLU • APPENDIX 11 (DOC 61) AL 3/24/19 326485

housing programs, including but not limited to all HUD-funded programs. I have been employed by the City of Providence since 2016. 3. As the City’s Community Development Director, I develop strategic goals, solicit and evaluate project proposals annually, develop program budgets and spending plans, issue and evaluate federally-funded contracts, and report to HUD and other relevant funding entities. Part of my work entails working with federal grant programs, including Community Development Block Grant funds, Emergency Solution Grant funds, and HOME Investment Partnership Program Funds. 4. The City of Providence receives millions of dollars in federal funding from the U.S. Depart- ment of Housing and Urban Development (HUD) on both an annual and competitive basis. 5. Annual entitlement funding from the U.S. Department of Housing and Urban Development is allocated by funding formulas that include weighted population, poverty, housing counts and other data points. 6. In Federal Fiscal Year 2018, the City of Providence will receive $5,034,413 in Community Development Block Grant funding; $421,403 in Emergency Solutions Grant funding, and $1,700,757 in HOME Investment Partnerships Program funding. 7. The CDBG statute identifies poverty, deteriorated housing, economic distress, decline, and suitability of one’s living environment as important components of community development need. The City’s annual allocation of CDBG funds is calculated by HUD through Formula A or

78228 • ACLU • APPENDIX 11 (DOC 61) AL 3/24/19 327486

Formula B, as defined under 42 U.S.C. §5306. Formula A calculates funds to a jurisdiction based on its metropolitan area’s share of 3 factors across all US metropolitan areas: 1) population, weighted at 25 percent; 2) people in poverty, weighted at 50 percent; and 3) overcrowded housing units, weighted at 25 percent. Formula B also calculates funds to a jurisdiction based on its metropolitan area’ s share of 3 factors across all US metropolitan areas, but the Formula B factors are: 1) population growth lag since 1960, weighted at 20 percent; 2) people in poverty, weighted at 30 percent; and 3) pre-1940 housing units, weighted at 50 percent. After HUD runs the calculations using the two formulas, it allocates the City the larger amount of the two. In recent years, the City of Providence has been awarded funds under Formula B. A differential undercount on the 2020 Census that impacts the Providence metropolitan area more heavily than other metropolitan areas will result in underreporting of the population and population growth lag as those formula factors are based on last Decennial Census counts. 8. Further, to utilize CDBG funds on neighborhood facilities or improvements, neighborhood Census Block populations must be 51% or greater low/moderate income. Any undercounting of residents, particularly in neighborhoods with high poverty rates and large minority and immigrant populations, would result in fewer neighborhoods being eligible for these critical neighborhood-based investments. 9. Reduced CDBG funding will result in a reduction in critical social services (such as homeless and domestic violence case management,

78228 • ACLU • APPENDIX 11 (DOC 61) AL 3/24/19 328487

food pantry operations, and job training programs) as well as reduced investment in local infrastructure and affordable housing. If a 1% funding reduction were to occur, over $50,000 in critical services and programming would need to be cut from the City’s annual CDBG budget. Activities eliminated from a reduced City CDBG budget to absorb such a cut would be free daycare and after- school programs for hundreds of low-income families, and street and sidewalk improvements. 10. The Emergency Solutions Grant (ESG) is an annual entitlement grant program whereby funds are allocated under a 24 C.F.R. § 576.3-defined formula. Allocations are a function of the prior year’s CDBG allocation; therefore, if CDBG funds decrease, ESG funds will be reduced as well. ESG funds are used to provide rental assistance to persons experiencing homelessness, as well as street outreach and case management services. 11. A reduction in City ESG funding would result in higher rates of homelessness, displace- ment of at-risk populations from assisted housing, and a decline in case management and shelter services. Even a modest 1% reduction in funding ($4,000) would mean cutting hundreds of hours for local case workers-hours that would have been spent finding housing for homeless living and dying on Providence streets. 12. Eligibility of a local government to receive HOME Investment Partnerships Program (HOME) is also a factor of the amount of CDBG received by the municipality. The HOME formula is then further determined by the following factors: 1) local and national vacancy rates (sourced by the Decennial Census), 2) prevalence of rental housing

78228 • ACLU • APPENDIX 11 (DOC 61) AL 3/24/19 329488

with substandard conditions, 3) number of pre-1950 rental units (sourced by the Decennial Census), 4) RS Means cost index and substandard rental units, 5) number of families in poverty, and 6) low net per capita income and population (population figures from the Decennial Census). If a differential undercount on the 2020 Census results in more homes being identified vacant than actually are, the formula would be negatively impacted as local market tightness (a proxy for local need for more affordable rental housing units) would be understated. 13. A reduction in City HOME funding will result in a decrease in the production of new affordable housing and a reduction in local down- payment and closing cost assistance programs to aid first-time homebuyers with limited means. A reduction of $170,000 (1%) in HOME funds would result in approximately 22 low-income families not receiving down-payment and closing cost assistance and realizing the dream of homeownership. 14. Finally, it is important to note that all formula factors for the programs above are derived from either the Decennial Census or US Census Bureau American Community Surveys. I declare under penalty of perjury that, to the best of my knowledge, the foregoing is true and correct. Executed on this 24th day of October, 2018

/s/ EMILY FREEDMAN Director of Community Development for the City of Providence

78228 • ACLU • APPENDIX 11 (DOC 61) AL 3/24/19 330489

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

18-CV-2921 (JMF) ______

STATE OF NEW YORK, et al., Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE; et al., Defendants. ______

Affidavit of Dr. Hermann Habermann

I. Professional experience and qualifications. 1. I have over thirty-five years of experience as a statistician, earning much of that experience at statistical agencies of the United States government. Among other federal government positions, I have served as Chief Statistician of the United States (1988–1992) and as Deputy Director and Chief Operating Officer of the United States Census Bureau (2002–2007). I also served for eight years as Director of the United Nations Statistics Division (1994–2002).

78228 • ACLU • APPENDIX 12 (DOC 62) AL 3/24/19 331490

2. I earned my Ph.D. in Statistics from the University of Wisconsin–Madison in 1975. A copy of my curriculum vitae is PX-353. II. Summary of Findings. 3. I was retained by the plaintiffs in this litigation to provide my expert opinion on the policies and procedures federal statistical agencies follow when designing, modifying, and implementing statistical instruments, and on the extent to which the Commerce Department and *** the 2020 Census, none of those processes were followed with respect to adding the citizenship question to the 2020 Decennial Census short form. 40. Despite repeated references in the Admin- istrative Record to steps involving “robust processes” for working with OMB and the ICSP-SACS [see, for example, PX-004 (AR 003890), PX-004 (AR 005567), PX-004 (AR 005512)], the addition of the citizenship question just prior to the submission of planned questions to the Congress appears to have taken place without any apparent consultation with OMB or the ICSP-SACS. In fact, as of March 6, 2018, a Census Bureau briefing for the Department of Commerce indicated that there would be no changes to the 2020 Census subjects, that an OMB briefing had taken place on February 22, and that the ICSP-SACS briefing would take place on March 14. PX-001 (AR 000435). 41. In addition, as noted above, the Census Bureau requested a meeting with the Department of Justice in order to give the technical experts an opportunity to discuss the details of the proposal to

78228 • ACLU • APPENDIX 12 (DOC 62) AL 3/24/19 332491

add a citizenship question, but the Department of Justice declined to meet. PX-004 (AR 009074); (Abowd 8/30/2018 dep. tr. 96-99). As Dr. Abowd testified on behalf of the Census Bureau, it is unusual to receive a data request from an agency and then for the agency to refuse to meet to discuss the technical aspects of that data request. (Abowd 8/30/2018 dep. tr. 98-99). 42. The content review process for the DOJ’s request for a citizenship question did not follow the basic protocols for interacting with OMB and the Census Bureau to assure that questions on the decennial census are required by federal programs and that the information collection is as minimally burdensome as possible. 43. An example of a typical interagency collabo- ration process illustrates the extent to which the Commerce Department deviated from typical practice here. Revisions to survey questions in the federal statistical system usually involve an extensive, multi-faceted process. This is particularly true when a question may be employed, or its data used, by multiple agencies. A salient case is the question employed across the government when asking about race and ethnicity. Prior to the mid- 1970s, there was no standard approach to asking this question on federal information collections, whether for general demographic information, for the evaluation of federal program initiatives, or for enforcement of government policies. At the request of several agencies, OMB undertook the development of a standard approach to collecting this information, and in 1977 issued a standard for use by agencies that intended to collect race and ethnicity data. For the first time, the “denomi-

78228 • ACLU • APPENDIX 12 (DOC 62) AL 3/24/19 333492

nators” from the decennial census could be used with “numerators” from various surveys (e.g., education, labor, health) as well as administrative reports (school enrollment, employee characteristics, patient records) to better understand access to learning, labor force participation, and use of services. The 1977 standard, based largely on the question then used by the Office of Education in the Department of Health, Education, and Welfare, was adopted following an OMB-led consultation with several federal agencies. 44. In the early 1990s, Congress highlighted the need for a review and possible revision of the OMB standards for data on race and ethnicity. OMB agreed to undertake this review, outlining a three- pronged process. This included: (1) establishing an interagency committee comprising the producers and users of data on race and ethnicity (30-plus agencies); (2) conducting a research and testing program to examine and assess alternatives that were under consideration; and (3) providing multiple opportunities for public input and comment on options (via public hearings as well as multiple Federal Register notices). 45. The process ultimately spanned four years, from inception to announcement of revisions (October 1997). During that time, both affected federal agencies and stakeholders outside government had multiple, continuing opportunities to contribute to the research agenda and to comment on the potential changes. The incremental research and testing program (which included substantial cognitive work) allowed for full consideration of alternatives, some of which came into view as the process unfolded.

78228 • ACLU • APPENDIX 12 (DOC 62) AL 3/24/19 334493

46. The standards that ultimately were adopted, though not necessarily the first choice of some constituencies, were well-received as a consequence of the robust process that had been employed. A four-year process is not always necessary, but the critical components of the process need to be carried out: interagency involvement, research and testing, and timely public comment.

VI. The Commerce Secretary rejected a less- costly and better-quality alternative that the Census Bureau proposed for producing block level citizenship data. 47. Part IV of this affidavit discussed the standards that apply to an agency’s request for a particular data collection, including the requirement that the request be supported by sufficient justification to demonstrate the practical utility of a collection, and the necessity of that collection to properly perform a given agency function. I further discussed that insufficient information was provided in the Commerce Secretary’s March 2018 memo for rejecting the Census Bureau’s opinion that the information needed by the DOJ could be obtained through modeling of existing ACS data. 48. However, assume for the moment that such evidence had been supplied by DOJ. In that event, the Census Bureau did develop a solution which would provide block-level data on citizenship, but would not require adding a question on citizenship. Moreover, this option would, in the judgment of the Census Bureau, be less costly and provide better quality data than

78228 • ACLU • APPENDIX 12 (DOC 62) AL 3/24/19 335494

*** planning standard requires agencies to provide a justification that includes, in part, the decisions the survey is designed to inform, the precision required of estimates (e.g., the size of differences that need to be detected), when and how frequently users need the data, and the tabulations and analytic results that will inform decisions and other uses. The pretesting survey systems’ standard requires agencies to ensure that all components of a survey function as intended when implemented in the full- scale survey, and that measurement error is controlled by conducting a pretest of the survey components. 57. Standards and guidelines for cognitive interviews issued by OMB similarly apply to federal censuses and surveys. Cognitive interviewing is a key method used to pretest survey questions and questionnaires that can indicate whether a survey question captures the intended construct, and identify difficulties that respondents experience in understanding and accurately answering proposed questions. 58. In adding a question to a survey, the normal practice is to test the question. This testing is done to understand, inter alia, how the question will be received by different respondents (including response rates and quality of responses); what wording of the question performs best, and the question’s impact on other questions; and the correct placement of the question. These practices are part of the OMB standards. 59. With respect to these OMB standards on development of questions, the administrative

78228 • ACLU • APPENDIX 12 (DOC 62) AL 3/24/19 336495

record produced in this lawsuit shows that the Census Bureau stated it needed to undertake a rigorous process to evaluate proposed content additions to the decennial census. PX-004 (AR 003890). This process includes several steps related to testing. 60. For example, the Census Bureau wrote that it “must test the wording of the new question.” The Census Bureau stated that since “it is too late to add a question to the 2018 End-to-End Census Test . . . additional testing on a smaller scale would need to be developed and implemented as soon as possible. This test would also require approval from OMB, which includes notifying the public and inviting comments through a Federal Register Notice (FRN).” 61. In addition, the Census Bureau wrote that it “must make additional operational adjustments, beyond testing, to include new content. This includes re-designing the paper questionnaires and adjusting the paper data capture system. For all automated data collection instruments (including Internet self-response, Census Questionnaire Assis- tance, and Nonresponse Follow up), the additional question will require system redevelopment, for English and all supported non-English languages. In addition, the training for the enumerators and Census Questionnaire Assistance agents will need redevelopment.” 62. The Census Bureau also explained that “[b]ased on the result of the testing, the Census Bureau must finalize the actual 2020 Census questionnaires (paper and automated). The Census Bureau then must submit for OMB approval of the

78228 • ACLU • APPENDIX 12 (DOC 62) AL 3/24/19 337496

2020 Census information collection. This submission also requires notifying the public and inviting comments through a Federal Register Notice.” 63. This rigorous and “well-established process” is referenced repeatedly in the Administrative Record and is consistent with OMB standards. PX- 004 (AR 004773; AR 004874; AR 005512; AR 005565; AR 005567). 64. In his March 26 memo, the Secretary, while acknowledging the principle of testing, concludes: “[t]he Census Bureau staff have advised that the costs of preparing and adding the question would be minimal due in large part to the fact that citizenship question is already included on the ACS, and thus the citizenship question has already undergone the cognitive research and questionnaire testing required for new questions.” PX-001 (AR 001313).

78228 • ACLU • APPENDIX 12 (DOC 62) AL 3/24/19 338497

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

CIVIL ACTION NO. 1:18-cv-2921 (JMF) ______

STATE OF NEW YORK, et al., Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE; et al., Defendants. ______

AFFIDAVIT OF KATHERINE HARVELL HANEY

Pursuant to 28 U.S.C. § 1746, I, Katherine Harvell Haney, do hereby depose and state the following: 1. I am the Chief Financial Officer at the Massachusetts Executive Office of Health and Human Services. I have served in this role for the last eleven months. Previous to serving in this role, I spent ten years working in various finance capacities for Masshealth, the MA Department of

78228 • ACLU • APPENDIX 13 (DOC 63) AL 3/24/19 339498

Elder Affairs and the MA department of Mental Health. I also spent five years as the Budget Director for the Harvard T.H. Chan School of Public Health managing the school’s more than $120,000,000 annual federal grant expenditures. 2. I have either personal knowledge of the matters set forth below or, with respect to those matters for which I do not have personal knowledge, I have reviewed information gathered for me in my capacity as Chief Financial Officer. 3. The Executive Office of Health and Human Services (EOHHS) is the secretariat responsible for approximately fifteen agencies that provide benefits and services to some of Massachusetts’ most vulnerable citizens, including children, families, individuals with disabilities and elders. EOHHS works with the agencies within EOHHS to ensure federal grant requirements such as reporting, state plan compliance and federal match requirements are met so that agencies may draw down federal revenue to cover allowable grant expenditures. 4. Many of the agencies within EOHHS rely on federal grants to fund the services they provide to these vulnerable populations. Many of these federal grants use state population figures from the federal decennial census to determine the amount of funding available under the grant The Department of Children and Families, the Massachusetts Commission for the Blind, the Department of Public Health, the Department of Mental Health, the Massachusetts Rehabilitation Commission, and the Executive Office of Elder Affairs together receive fourteen different grants that consider state population figures from the decennial census to

78228 • ACLU • APPENDIX 13 (DOC 63) AL 3/24/19 340499

determine the amount of funding allocated to the Commonwealth. 5. For example, the Massachusetts Rehabilitation Commission (MRC) and the Massachusetts Com- mission for the Blind (MCB) receive funding from the Vocational Rehabilitation State Grant (VR grant). The VR grant provides funding to state agencies that offer vocational rehabilitation (in Massachusetts, MRC and MCB are the agencies that receive VR grant funds) to pay for services for individuals with disabilities to prepare for and engage in employment with the ultimate goal of achieving economic self-sufficiency. The formula for distributing VR grants to states and territories is to first allocate what the state or territory received in fiscal year 1978. Of the remainder of the funds, one-half is distributed based upon states’ general population, as established by the decennial census, and a factor that compares the state’s per capita income to the national per capita income; the other one-half is distributed according to the state population, again drawn from decennial census figures, and the square of the per capita income. The larger a state’s population, the more funds it will receive, but the higher a state’s per capita income compared to the national level, the lower its allotment will be. In this way, an undercount of low-income residents of a state would have a heightened impact on reducing VR grant funding received by that state. 6. In State Fiscal Year 2018 (SFY18), Massa- chusetts received approximately $46 million in VR grant funds. If Massachusetts’ share of the population relative to other states were to appear to decrease because of an undercount in the decennial

78228 • ACLU • APPENDIX 13 (DOC 63) AL 3/24/19 341500

census, the formula used to distribute VR grant funds to states would award less to Massachusetts. 7. Similarly, Community-Based Child Abuse Prevention (CBCAP) Grants are given to state child welfare agencies to support community-based efforts to prevent child abuse and neglect. In SFY18, the Massachusetts Department of Children and Families received approximately $460,000 in CBCAP funding. Funds are distributed to states based on the number of children under age 18 in each state, which figure is drawn from decennial census data. If the population of children under 18 in Massachusetts were to appear to decrease relative to other states because of an undercount in the census, the formula used to distribute the CBCAP Grants would award less to Massachusetts. In this way, even if there is little to no net undercounting of the national population, if the Massa- chusetts population of children under 18 were to be disproportionately undercounted as compared to other states, the Massachusetts’ percentage of all children under 18 in the nation would appear to decrease, and Massachusetts’ share of CBCAP grants would also decrease. 8. Finally, the Older Americans Act (OAA) Grants for State and Community Programs on Aging are given to state agencies designated as “State Units on Aging” and used to support the state’s over 60 population to fund home and community-based services, such as home-delivered meal and other nutrition programs, in-home services, transportation, legal services, elder abuse prevention and caregiver support. In SFY18, the Executive Office of Elder Affairs received approximately $26 million in OAA funding. Each

78228 • ACLU • APPENDIX 13 (DOC 63) AL 3/24/19 342501

state receives OAA funds according to a formula based on the state’s share of the national population of seniors, age 60 and older, as assessed by the decennial census. If the population of 60 and older residents in Massachusetts were to appear to decrease relative to other states’ populations of residents aged 60 and older because of an undercount in the decennial census, the formula used to distribute OAA funds to states would award less to Massachusetts. 9. If there is an undercount of eligible populations in Massachusetts as compared to other states, Massachusetts would receive less funding even though the actual number of individuals in the eligible populations has not decreased. I declare under penalty of perjury that the foregoing is true and correct. Executed on October 24, 2018. /s/ Katherine Harvell Haney Chief Financial Officer Executive Office of Health and Human Services Commonwealth of Massachusetts

78228 • ACLU • APPENDIX 13 (DOC 63) AL 3/24/19 343502

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

Civil Action No. 1:18-cv-2921 (JMF) ______

STATE OF NEW YORK, et al., Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE, et al., Defendants. ______

AFFIDAVIT OF JASON HARMON Pursuant to 28 U.S.C. § 1746(2), I, JASON HARMON, hereby declare as follows: 1. I am over the age of eighteen and have personal knowledge of all the facts stated herein. 2. I am the Director of the Office of the Every Student Succeeds Act (“ESSA”) Funded Programs at the New York State Education Department (“NYSED”) and have been employed by NYSED since 2013. Part of my work includes overseeing the federally-funded programs described below.

78228 • ACLU • APPENDIX part: 64 NP 5:40 3/24/19 344503

3. My office administers approximately $1.4 billion in federally funded ESSA Programs to Local Education Agencies (“LEAs”), Boards of Cooperative Educational Services (“BOCES”), and Institutions of Higher Learning (“IHL”) across the State each year, and is responsible for monitoring the programmatic and fiscal components for each of the federally funded programs identified below: o Title I, Part A - Improving basic programs operated by LEAs; o Title I, Part C - Education of migratory children; o Title I, Part D - Prevention and intervention programs for children and youth who are neglected, delinquent, or at- risk; o Title I - School Improvement Grants (“SIG”) o Title II, Part A - Supporting effective instruction; o Title IV Part A - Student Support and Academic Enrichment Grants; o Title V, Part B - Rural education; and o Title IX - McKinney-Vento Homeless Assistance Act. 4. Many of the ESSA funded programs utilize formulas that directly rely on the population counts for children living in poverty (“children counts”) from the federal Census Bureau. Any decline in the number of children in these counts would have a significant impact on not only State resources, but on the ability of school districts and charter schools to serve their administrators,

78228 • ACLU • APPENDIX part: 64 NP 5:40 3/24/19 345504

teachers and more importantly, the students of this State. 5. For example, Title I, Part A funds of the Elementary and Secondary Education Act (ESEA), as amended by ESSA, are used to provide financial assistance to LEAs and schools with high numbers or high percentages of children in poverty to help ensure that all children meet challenging state academic standards. The ESSA statute requires the U.S. Department of Education (“USDOE”) to calculate Title I allocations based on four formulas: Basic Grants, Concentration Grants, Targeted Grants, and Education Finance Incentive Grants (“EFIG”) to LEAs. Each of these grants directly utilizes the definition of “formula children” to determine each LEA’s allocation. The definition of formula children is based on annually-updated poverty estimates by the federal Census Bureau and this count affects the amount of funds an eligible LEA receives under each formula. 6. For the basic grants formula, an LEA is eligible for funds if the LEA has at least 10 formula children and that number exceeds two percent of the Census Bureau’s estimate of the LEA’s ages 5 to 17 population. An LEA is eligible for concentration grants, if the LEA meets the basic grant’s eligibility formula and if its number of formula children exceeds 6,500 or 15 percent of the Census Bureau’s estimate of the LEA’s ages 5 to 17 population. An LEA is eligible for Targeted Grants and EFIG if the LEA has at least 10 formula children and that number equals or

78228 • ACLU • APPENDIX part: 64 NP 5:40 3/24/19 346505

exceeds five percent of the Census Bureau’s estimate of the LEA’s ages 5 to 17 population. 7. New York State’s 2017-2018 Title I, Part A allocation was $1,208,000,088 and its 2018-2019 allocation is $1,213,935,615. 8. Title I, Part A funds are used by LEAs for academic support for at-risk students, including, but not limited to: academic intervention services (AIS) and response to intervention (RTI) services provided by certified teachers and teaching assistants; counseling, school-based mental health programs, specialized instructional support services, mentoring services supported by school counselors, social workers, and other support personnel; prekindergarten programs; summer and extended day programs; professional development for teachers, teaching assistants, and other school personnel; and equitable services to students in participating private schools. 9. These programs and services for at-risk students would be directly and substantially impacted by any decrease in the count of “formula children” and the resulting decrease in Title I, Part A funds. For example, a decrease in the count of “formula children” as small as 0.02 percent could cause an LEA to lose funding under the Targeted Grants and EFIG. Among LEAs that lost funding under those two programs in the 2018- 2019 school year, the average decrease was nearly 40 percent of Title I, Part A funding compared to the previous year. For one LEA, a decrease of 1.54 percent in the count of “formula children” resulted in a loss of nearly $300,000 in Title I, Part A funding in a single year. For an LEA with an

78228 • ACLU • APPENDIX part: 64 NP 5:40 3/24/19 347506

average teacher salary of $50,000 when budgeting, losing $300,000 translates into a loss of approximately 6 full time academic support teachers, and potentially threatens hundreds of students with a reduction in services. 10. Title II, Part A funds from ESEA, as amended by ESSA, are also directly reliant on poverty counts from the federal Census Bureau data. There is a two-step allocation process for Title II, Part A funds. First, the USDOE calculates an SEA apportionment, using the most recent federal Census Bureau data. For Fiscal Year (FY) 2017, the amount of funds allocated to each state was based on the following: • 35 percent according to each State’s population of children ages 5 through 17 relative to the number of these children in all States; and • 65 percent according to each State’s relative numbers of individuals ages 5 through 17 from families with incomes below the poverty line relative to the number of these children in all States. 11. Beginning in fiscal year 2018, new percentages have been phased in. The share of excess funds allocated on the basis of a State’s relative number of children ages 5 through 17 from families with incomes below the poverty line increases and the share allocated on the basis of a State’s relative number of children ages 5 through 17 decreases, as follows:

78228 • ACLU • APPENDIX part: 64 NP 5:40 3/24/19 348507

Fiscal Year Percentage based Percentage based on population on population ages 5 - 17 ages 5 - 17 in poverty FY 2017 35 65

FY 2018 30 70

FY 2019 25 75

FY 2020 28 80 and subsequent years 12. After required reserves for subgrants to eligible LEAs and allowable reserves for administration and state-level activities, the SEA, then distributes funds to LEAs based solely on the following formula: • 20 percent of the funds must be distributed to LEAs based on the relative numbers of individuals ages 5 through 17 who reside in the area the LEA serves (based on the most recent Census data, as determined by the Secretary); and

78228 • ACLU • APPENDIX part: 64 NP 5:40 3/24/19 349508

• 80 percent of the funds must be distributed to LEAs based on the relative numbers of individuals ages 5 through 17 who reside in the area the LEA serves and who are from families with incomes below the poverty line (based on the most recent Census data, as determined by the Secretary) (ESEA section 2102(a)). 13. Both funding formulas directly rely on population counts from the federal Census Bureau. If the population counts for children under the Census Bureau were to decline, this would have a negative direct impact on not only statewide activities supported by the New York State Education Department, but on the nearly 1,000 school districts and charter schools across New York State. 14. New York State’s 2017-2018 Title II, Part A allocation was $166,390,917 and its 2018-2019 allocation is $157,360,601. 15. Specifically, a decrease in Title I, Part A and Title II, Part A funds would immediately reduce the amount of available administrative funds used to support State-level program activities. Such a reduction would immediately and substantially diminish NYSED’s capacity to: assist schools, districts, charter schools, and private schools with basic program implementation; develop tools and resources such as technical assistance aids, guidance materials, and other supports; and conduct effective compliance monitoring based on established risk protocols. A decrease in Title II, Part A funds would also

78228 • ACLU • APPENDIX part: 64 NP 5:40 3/24/19 350509

drastically reduce the State’s capacity to provide guidance and resources on the Next Generation Learning Standards; to train teachers in low performing schools on how to develop curriculum materials aligned to such standards; and to ensure that teachers properly prepared on these new standards and appropriately qualified to teach low income students. These supports and training are necessary to ensure that low income students will have consistent or equitable access to effective educators and properly trained and skilled teachers as compared to their peers from lower- needs communities across the state. 16. Beyond the negative impacts described above, decreases in Title I, Part A funding as the result of a decline in population counts used in the federal Census Bureau data would also have a disproportionately negative impact on the lowest performing schools and districts in the state. Under Section 1003 of ESSA, 7 percent of Title I funds must be set-aside annually for the purposes of School Improvement. School Improvement Grants are funds that are set aside specifically for the lowest achieving schools, that have the greatest need for funding and who demonstrate the strongest commitment to ensuring funds are used to improve the academic achievement of the lowest performing students in these schools. Low performing LEAs use School Improvement funds help implement critical supports, services, and evidence-based interventions detailed in school and district-level improvement plans that are informed by a state-required review process. Funds to support improvement plans are

78228 • ACLU • APPENDIX part: 64 NP 5:40 3/24/19 351510

primarily targeted at addressing (1) the specific academic, social-emotional, and safety needs of students, (2) the professional development needs of teachers and (3) effectively engaging parents and families. 17. An LEA’s funding allocation for Title IV, Part A of ESSA is also contingent upon the LEA’s funding allocation provided under Title I, Part A, which as described in paragraph 5 above, is directly reliant on the population counts of the federal Census Bureau. As a result, if an LEA’s Title I allocation decreases as a result of a decline in child poverty counts, its Title IV, Part A allocation will also decrease. Title IV, Part A funding is used to support well-rounded educational programs such as college and career counseling, STEM education, arts, civics and advanced programs such as International Baccalaureate/Advanced Placement; safe and healthy school programs including school climate surveys, comprehensive school mental health, drug and violence prevention, training on trauma- informed practices, and health and physical education; and to support the effective use of technology. These services may be at risk with any decrease in funding. I declare under penalty of perjury that, to the best of my knowledge, the foregoing is true and correct. Executed on this 23rd day of October, 2018 /s/ JASON HARMON

78228 • ACLU • APPENDIX part: 64 NP 5:40 3/24/19 352511

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

Civil Action No. 1:18-cv-2921-JMF ______

STATE OF NEW YORK, et al., Plaintiffs,

NEW YORK IMMIGRATION COALITION, et al., Consolidated Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE, et al., Defendants.

______

DECLARATION OF SAMER E. KHALAF

I, Samer E. Khalaf, pursuant to the provisions of 28 U.S.C. § 1746, declare under penalty of perjury that the foregoing is true and correct: 1. I am the National President of the American-Arab Anti-Discrimination Committee (“ADC”) and ADC Research Institute (“ADCRI”).

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 353512

My responsibilities include overseeing all of the work of both organizations, and leading our efforts regarding the 2020 Decennial Census. 2. Based on my knowledge as a leader in the Arab-American community, ADC and ADCRI’s community outreach work, and my interactions with individual ADC members, I believe the addition of a citizenship question to the 2020 Decennial Census short-form questionnaire will deter participation of many individuals in the Arab-American community and result in a disproportionate undercount of Arab Americans. As a result of the decision to add this question, ADC and ADCRI have diverted and will continue to be required to divert resources from core organizational priorities to additional, remedial outreach to counteract the citizenship question’s negative effect on Census response rates in our community. 3. This undercount will also injure many ADC members, many of whom live in areas of the United States in which immigrants of color and other communities of color exceed national and state averages. Specifically, they will be harmed because the disproportionate undercount that will result from the citizenship question among immigrants of color will diminish their political power relative to other parts of their states as well as the amount of Census-related funding those areas receive for key programs, services, and facilities that our members and their communities use.

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 354513

I. Background on ADC and ADCRI

A. ADC and ADCRl’s Mission and Activities 4. Senator James G. Abourezk founded ADC in 1980. As the first Arab American to serve in the United States Senate, Abourezk founded ADC in response to stereotyping, defamation, and discrimination directed against Americans of Arab origin. 5. In 1981, Senator Abourezk founded ADCRI. It is a corporation exempt from taxation under Section 501(c)(3) of the Internal Revenue Code. The Institute sponsors a wide range of programs on behalf of Arab Americans and of consequence to the wider American community. 6. ADC is a civil rights membership organization that is committed to defending and promoting the rights and liberties of Arab Americans and other persons of Arab heritage. In fact, ADC is the largest American-Arab grassroots civil rights organization in the United States. 7. ADC’s mission focuses on combating stereotypes and discrimination against and affecting the Arab-American community in the United States and serving as its public voice for on domestic and foreign policy issues, as well as educating the American public in order to promote greater understanding of Arab history and culture. ADC advocates, educates, and organizes to defend and promote human rights and civil liberties of Arab Americans and other persons of

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 355514

Arab heritage, from recent immigrants to citizens born in the United States. 8. ADC has several thousand members who reside in all 50 states and the District of Columbia. Many of these members are active in ADC’s 28 local chapters, which are located in 20 states and the District of Columbia. These include active chapters with members leading local efforts, including in Tucson and Phoenix, Arizona; Los Angeles and Orange County, California; Miami and Orlando, Florida; New York, New York; and Austin and Dallas, Texas. 9. ADC focuses its legal advocacy efforts on issues that have a large impact on the Arab American community, which includes employment discrimination, education discrimination, and immigration. The Legal Department has continued its efforts to work with established coalitions and build new partnerships with other minority groups on issues of mutual interest. Furthermore, the Legal Department maintains a strong relationship with government agencies to ensure that the diverse voices of the Arab American community are heard by key decision- makers in American government. 10. ADCRI’s programs train Arab Americans in the exercise of their constitutional rights as citizens. They include research studies and publications, seminars, and conferences, which document and analyze the discrimination faced by Arab Americans in the workplace, schools, media and government agencies. They also promote a better understanding of the rich Arab cultural

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 356515

heritage by the public, policy makers, and the new generations of Arab Americans. 11. A major activity of ADCRI is the Anthony Shadid Internship Program, which annually brings undergraduate, graduate, and law students to Washington for hands-on experience in the Nation’s capital. The program provides them the opportunity to learn about issues confronting Arab Americans, to develop effective leadership skills, to gain experience in research and writing, and to develop greater knowledge and understanding of the complexities of the U.S. government and our political process. 12. ADCRI programs include the publication and distribution of educational materials to Arab Americans, educators, academic institutions, policy makers, and the public at large. The Institute has in the past and will continue to foster intergroup harmony in American society, the entry of Arab Americans into the mainstream of American life and leadership, and the abolition of prejudice, stereotyping, defamation, hate crimes, and violence directed at the Arab American and Muslim communities.

B. ADC and ADCRI’s History of Census- Related Work 13. ADC and ADCRI have an ongoing commitment to promoting engagement in the Decennial Census among its members and constituents. Since the 1980s, ADC has served in numerous capacities on the Census Bureau advisory committees. In 2000, ADC reinvigorated its census campaign and policy initiatives, actively

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 357516

educating Arab Americans and other minority populations about the importance of the census and mobilizing participation and outreach events. In 2008, ADC actively worked with the Census Bureau to identify classification concerns of Arab Americans on the United States census. 14. In the months leading up to and during the 2010 Decennial Census, ADC and ADCRI conducted outreach and engagement work with the Arab-American community concerning census engagement. That work consisted of creating messaging about participating in the Decennial Census that was focused on the Arab-American community including launching a national campaign advocating for the completion of census forms and holding outreach events to counter concerns of sharing of information with government and law enforcement agencies. ADC and ADCRI also engaged in efforts to “get out the count” in that community as well, including hosting town halls and symposiums in select cities across the country. ADC and ADCRI also produced and published material for distribution which encouraged participation in the 2010 Census and highlighted its importance. ADC also served on the U.S. Census Bureau’s 2010 Census Advisory Committee. II. Organizational Harm to ADC and ADCRI as a Result of the Citizenship Question 15. Because of the importance of having Arab Americans accurately counted in the 2020 decennial census, ADC and ADCRI had already begun preparations for outreach to the Arab- American community before the announcement of

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 358517

a citizenship question. Specifically, in 2017, ADC and ADCRI had begun focusing on educating community members about the Decennial Census and its importance. The announcement in late March 2018 that the 2020 Decennial Census would include a citizenship question, however, changed the plans of both ADC and ADCRI. 16. As described below, the decision stoked and continues to stoke tremendous fear in the Arab- American community and threatens to lower response rates and increase the undercount of our community. As a result, this decision has harmed part of ADC and ADCRI’s organizational missions and forced them to divert resources from other programmatic attempts to ameliorate some of the citizenship question’s impact.

A. Fear in the Arab-American Community Due to the Citizenship Question and its Impact 17. For a number of reasons, the Arab-American community has traditionally been significantly undercounted in the decennial census. Because of this, before the announcement of the citizenship question, ADC and ADCRI already saw a need to conduct census outreach. Additionally, some outreach was already planned because of ADC members’ and the Arab-American community’s heightened fear of interacting with government workers due in part to the Trump Administration’s persistently hostile and discriminatory actions and attitudes towards Arab Americans, Muslim Americans, and immigrants of color generally.

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 359518

18. Those fears, however, and the risk of an even greater undercount in the Arab-American community, have already been greatly exacerbated by the Administration’s announcement that it will add a citizenship question and will be further exacerbated if the citizenship question is ultimately included in the 2020 Census. This fear is not unique to undocumented immigrants or non-citizens with documented legal status, but also to family and household members of non-citizens and their friends and neighbors, who have expressed concern that participating in the Census might endanger their loved ones. 19. Specifically, I know based on interactions with ADC members and people in the communities they serve that some Arab Americans would be fearful of responding and may not respond to the Census questionnaire if the citizenship question is added. ADC hosted different town hall events throughout the country, and Census 2020 was covered. I learned from those sessions that many community members fear the addition of the citizenship question and are unwilling to participate in the 2020 Census out of the fear deriving from the citizenship question. 20. For example, even in Dearborn, Michigan, which has the highest concentration of Arab Americans in the United States, many community members fear responding to the Decennial Census because of the existence of the citizenship question and the improper purposes for which they believe it might be used against them and their neighbors. Even with outreach from ADC

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 360519

and active leaders in the organization such as Amer Zahr in Dearborn, many in that community are not able to be persuaded to participate in the 2020 Census largely because of the citizenship question. From ADC events held at a number of locations throughout the country; interactions with individual ADC members; our various “Know Your Rights” seminars in several locations including Dearborn, Buffalo, Queens, Brooklyn, and Minnesota; and information gained at our 2018 National Convention, I am aware of similar concerns among community members in New York, Texas, Michigan, and California. 21. Specifically, Arab American communities fear that the citizenship question will cause community targeting by ICE and DHS. Community members fear that the citizenship census data would be utilized for deportation and government surveillance purposes. Similarly to how the Census Bureau utilized its data to help the government round up Japanese Americans into internment camps during World War II, Arab Americans fear that the Census Bureau would report areas with high concentrations of noncitizen Arab Americans to the government for the purposes of deportation, censorship, and surveillance. In areas like Dearborn or Orange County, California, which contain high concentrations of Arab Americans, there is a fear that the citizenship question would lead to irreversible DHS targeting of the community. 22. After meeting with community members across the country, it is clear that—due in significant part to the citizenship question and the

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 361520

context in which this Administration added it— the community fears that census participation will result in loss of citizenship, revocation of LPR status, or the broadening of Presidential Proclamation 9645, which implemented the travel ban on several Arab- and Muslim-majority countries. The community is hesitant to participate in the Census due to the citizenship question and the possible effects it can have on the immigration status of individuals, families, friends, and other community members.

B. Harm to ADC and ADCRI from the Citizenship Question 23. The fear and likelihood of higher non- participation in the Decennial Census by Arab Americans due to the announcement of the citizenship question and the ultimate presence of the question on the Decennial Census questionnaire has harmed the ability of ADC and ADCRI to carry out key aspects of their missions. It has also forced them to divert resources from other programmatic areas to try to encourage Arab Americans to participate. 24. Despite these efforts, ADC is already facing a much more challenging environment for conducting outreach and encouraging Arab Americans, both citizens and noncitizens, to respond to the 2020 Census than the 2010 Census. Because of the heightened fear and suspicion created by the citizenship question, ADC has recognized the need to expand upon the census- related services, programming, and support it had already to try to reduce the negative effect of this

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 362521

question on the response rate in the Arab- American community. 25. The citizenship question has posed greater challenges for ADC and ADCRI to engage with the Arab American community on Census 2020 related matters. In 2010, and prior Census outreach efforts, ADC and ADCRI found it easier to work with the community on pushing the importance of Census participation. The citizenship question adds a new dynamic of fear and uncertainty. Additional outreach, meetings, and informational sessions are being planned and implemented in an effort to answer questions and address concerns. This year, ADC and ADCRI will hold focus groups to test messaging about Decennial Census participation and will follow up with polling afterwards. As the Decennial Census draws nearer, ADC and ADCRI will conduct training for census enumerators, run advertise- ments encouraging participation, and hold a strategy symposium, among other activities. ADC and ADCRI will also host community symposiums across the country in an effort to address concerns and answer questions. In addition, ADC and ADCRI will be working with coalition partners on issues of mutual concern relating to the 2020 Census, including the citizenship question. 26. Specifically, ADC and ADCRI have put additional resources toward a number of census- related efforts in large part because of the need to try to counteract the expected effects of the citizenship question on Arab Americans. As one example, on October 12, 2018, as part of its annual National Convention, ADC hosted a panel titled: “We Count Too! A Look at the 2020

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 363522

Census.” See PX-251.30. One of the primary purposes of, and topics covered during, that Panel was the addition of the citizenship question to the Decennial census form. In addition to offering their own thoughts about the citizenship question, the three panelists fielded questions from ADC members in the audience that reflected fears and concerns about the role of this question on the census. 27. ADC and ADCRI have also put an increased amount of time and effort into work with members of multiple coalitions to address the likely impact of the citizenship question, including the Leadership Conference on Civil and Human Rights and the Funders’ Committee for Civic Participation’s Census Initiative 2020. This work includes advocacy on Capitol Hill, with the Census Department, Commerce Department, and the Office of Management and Budget. Additionally, we have worked to put together new strategies to educate and communicate with our community. 28. ADC and ADCRI expects that they will need to interact with its constituents multiple times to answer questions and try to convince them to participate in the 2020 census, spending more resources to reach the same number of people. Despite this, they expect that it will be less successful in convincing its constituents to participate in the 2020 Census than in 2010 due in large part to the presence of the citizenship question. Collectively, ADC and ADCRI anticipate spending at least $150,000 more on 2020 Census education and outreach that the organization spent on its efforts to encourage participation in

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 364523

the 2010 Census as a result largely of the presence of the citizenship question. 29. Because of the need to increase the time and money spent on Census outreach due to the addition of the citizenship question, ADC and ADCRI will need to divert resources from other areas critical to its mission including organizing, issue advocacy efforts and educational initiatives. ADC and ADCRI have already diverted resources from these other areas in order to address concerns from their constituents stemming from the announcement of the citizenship question. As the largest Department in ADC, resources will be diverted from legal work to Census related matters. The diverting of resources means ADC will have less money to use towards assisting victims of hate crimes, and providing pro bono legal services. The ADC Legal Department is pro bono, thus the resources committed to the department are essential in providing services to community members across the country. III. Associational Harm to ADC through Harm to its Members as a Result of the Citizenship Question 30. ADC has members in all 50 states plus Washington, DC, as well as a national network of chapters. 31. ADC has a national network of chapters with members in all 50 states. PX-252. ADC has Board of Directors, most of whom membership elects, and all of whom are also ADC members and represent a cross-section of the United States, including: Los Angeles County, California (Dr.

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 365524

Souhail Toubia and Dr. Diane Shammas); San Diego, California (George Majeed Khoury); Brooklyn, New York (Dr. Debbie AlMontaser); Miami, Florida (Shatha Atiya); Washington, DC (Dr. Safa Rifka and Dr. Doo’a Taha); and Wayne, New Jersey (Abed Awad), among other cities and states. PX-253.6. 32. ADC also has 28 local chapters, which are located in 20 states and the District of Columbia. These include active chapters with members leading local efforts, including in Tucson and Phoenix, Arizona, Los Angeles and Orange County, California, Miami and Orlando, Florida, New York, New York, and Austin and Dallas, Texas. 33. For example, ADC member and Austin, Texas chapter President Ahmad Zamer has helped reinvigorate the Austin Chapter in recent years. In 2016 and 2017, the Austin chapter held a number of events, including a “know your rights” event, film screening, gala, and Iftar dinner for Austin’s Arab, Muslim, and Latino communities. PX-251.38, 39. Similarly, ADC has several active California chapters with dozens of members. The Orange County and San Diego chapters hosted a number of events, including a refugee empowerment event, a letter-writing event, and a memorial banquet. PX-251.38, 40. 34. These documents, as well as my personal knowledge, establish that ADC has members in all fifty states plus the District of Columbia. I have personal knowledge of each of these documents and attest to their authenticity.

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 366525

35. Many of ADC’s dues-paying members reside in states and localities in which the percentage of immigrants of color exceeds state and national averages. This includes: 172 in Arizona; 1,441 in California; 1,296 in D.C.; 551 in Florida; 437 in Illinois; 612 in Maryland; 819 in New York; 341 in Ohio; 1,341 in Pennsylvania; 30 in Rhode Island; 408 in Texas; and 186 in Washington State. In terms of local communities, ADC has members in communities with large concentrations of Arab- Americans such as Wayne County, Michigan, and Orange County, California, as well as communities with large Latino and noncitizen populations, including New York, New York, Los Angeles County, California, Houston, Texas, Miami, Florida, Prince George’s County, Maryland, Phoenix, Arizona, and Chicago, Illinois. A disproportionate undercount of those populations will deprive ADC members in those states and communities of political power by diluting the amount of state and congressional representatives to their areas. 36. ADC members across the nation rely on a number of facilities and services whose funding allocation relates to population and demographic data derived from the Decennial Census including parents with children enrolled in schools that receive funding under Title I of the Elementary and Secondary Education Act and drivers who use interstate highways and mass transit on a daily basis and thus depend on federal funds to perform their jobs.

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 367526

I declare under penalty of perjury that the foregoing is true and correct. Executed: October 26, 2018 Washington, DC /s/ Samer E. Khalaf

78228 • ACLU • APPENDIX part: 65 NP 5:10 3/24/19 368527

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

Civil Action No. 1:18-cv-2921 (JMF) ______

STATE OF NEW YORK, et al., Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE., et al., Defendants.

______

AFFIDAVIT OF CHRISTINE PIERCE

Pursuant to 28 U.S.C. § 1746(2), I, Christine Pierce, hereby declare as follows: 1. I am over the age of eighteen and have personal knowledge of all the facts stated herein. 2. I am the Senior Vice President of Data Science for The Nielsen Company (US) LLC (‘‘Nielsen”). I am a social scientist by training and worked as a demographer for Nielsen prior to my

78228 • ACLU • APPENDIX part:66 NP 4:51 3/24/19 369528

current role leading a team of scientists who support Nielsen’s audience measurement products. I earned a Master of Public Policy from the University of Minnesota and a graduate certificate in Applied Statistics from Pennsylvania State University. I frequently represent Nielsen at research conferences and have authored papers and presentations for the American Association of Public Opinion Research, the Advertising Research Foundation, and the Population Association of America. In the fall of 2018, the New York State Office of the Attorney General requested that I submit a voluntary affidavit in this case describing my communications with the Department of Commerce in lieu of potentially receiving a trial subpoena. 3. I am submitting this affidavit in order to ensure that the record accurately reflects my communications with the Department of Commerce. 4. In the spring of 2018, Nielsen received a request from an assistant to Secretary Ross asking to set up a meeting with someone at Nielsen who is familiar with Nielsen’s use of Census data. At the time, I was under the impression that the phone call would be to discuss the importance of the Census generally, the need for Nielsen and its commercial clients to have as complete and accurate a count as possible, and to advocate for full funding for Census operations. Nielsen’s SVP Community Engagement (Don Lowery) received this request. When Don Lowery sent the email connecting me to the Secretary’s office he included a statement that said “Christine

78228 • ACLU • APPENDIX part:66 NP 4:51 3/24/19 370529

looks forward to speaking to the Secretary regarding the importance of the 2020 Census to Nielsen.” See PX-532 (a true and accurate copy of email communications with Department of Commerce). 5. Prior to the phone call, Brian Lenihan from Secretary Ross’s staff asked me via email for a “copy of my biography (paragraph) along with a description of Data Science/Nielsen and how Census data comes into play.” I indicated that “For Nielsen, these public data sources such as the Decennial Census and ACS serve a crucial role in planning samples and consumer panels. Accurate population estimates enhances the sample design and ensures the most accurate coverage of households and persons with various demographic characteristics. Additionally, these public data sources are used to adjust the unweighted input to reflect the entire population.” See id. 6. I exchanged several emails with the staff regarding the date/time for the call. The staff did not mention the citizenship question in any of these emails. Other than the aforementioned biography and description, Secretary Ross’s staff did not ask me to provide any other documents or data nor did I provide any other data or documents to the Department of Commerce in Spring 2018. 7. On the evening of March 23, 2018, I had a telephone call with Secretary Ross and Michael Walsh, a lawyer from the Commerce Department. This telephone call lasted approximately 10-20 minutes. This was the only time that I spoke with

78228 • ACLU • APPENDIX part:66 NP 4:51 3/24/19 371530

Secretary Ross. I understand that three days after our conversation, Secretary Ross wrote a memo in which he discussed our conversation (the “Ross Memo”). 8. Prior to speaking to the Secretary, I was not aware that the citizenship question was going to be a topic of conversation. However, it immediately became apparent that the citizenship question was the only topic of conversation. Secretary Ross and Mr. Walsh told me that they needed to make a recommendation about whether to include a citizenship question on the Decennial Census and were reaching out to experts and stakeholders to gather information. 9. During this conversation, I told Secretary Ross unequivocally that I was concerned that a citizenship question would negatively impact self- response rates. I explained that people are less likely to respond to a survey that contains sensitive questions. I also added that increasing the length of a survey can reduce response rates. I discussed the impact that lower response rates have on survey costs. I emphasized that Census non-response follow up operations are expensive because they require a full count and nonresponse follow up operations for the Decennial Census include in-person data collection. 10. The Ross Memo states that I “confirmed that, to the best of [my] knowledge, no empirical data existed on the impact of a citizenship question on responses.” (Ross Memo at 3). I did not say “to the best of [my] knowledge no empirical data existed on the impact of a citizenship question on responses.” I did discuss

78228 • ACLU • APPENDIX part:66 NP 4:51 3/24/19 372531

the importance of testing questions to understand any impacts to response and I explained that a lack of testing could lead to poor survey results. I confirmed that I was not aware of any such test of a citizenship question by the Census Bureau. I cannot and did not attempt to quantify the extent of the reduction in self response. 11. During our conversation, Secretary Ross and Mr. Walsh asked me if Nielsen asked any sensitive questions. I told them that Nielsen does not ask about citizenship status on its surveys but that we do have surveys that occasionally include sensitive questions. 12. The Ross Memo explains that Nielsen “stated that it had added questions from the ACS on sensitive topics” including “immigration status to certain short survey forms without any appreciable decrease in response rates.” (Ross Memo at 3). I did not state that Nielsen had added “questions concerning immigration status to short survey forms without any appreciable decrease in response rates.” 13. I did explain to Secretary Ross and Mr. Walsh that Nielsen does ask certain questions from the ACS in our surveys and of our panelists, including place of birth and year of entry to the United States. I stressed the importance of specifically testing changes to questionnaires and that Nielsen had done such testing specifically because we anticipated these sensitive questions could have a negative impact on response rates. I did confirm that these place of birth and year of entry questions had not caused a significant decline in response rates on Nielsen surveys or in

78228 • ACLU • APPENDIX part:66 NP 4:51 3/24/19 373532

our panels. But I did not suggest that Secretary Ross could draw parallels between the surveys conducted by Nielsen and the Decennial Census. 14. Nielsen’s survey and panel operations are entirely different from the Decennial Census operations. Nielsen surveys are not conducted by a government agency and are not required by law. Nielsen studies are intended to understand consumer purchases and media usage. Response rates to the Nielsen surveys and panels in my purview generally range from 5% to 40%. If individuals do not answer a Nielsen survey or decline to participate in a panel, Nielsen will select and recruit different respondents to ensure we have the desired reporting sample size. While we strive for an accurate representation of the population, we are not required to count all people. And unlike the Census, Nielsen provides incentives — usually cash — for filling out our surveys. 15. To my knowledge, the Department of Commerce has not asked for any documents related to Nielsen’s survey work or questionnaire testing. To my knowledge, no one else at Nielsen has been asked for, or provided, any additional data, documents, or surveys to the Department of Commerce in response to the discussions around the citizenship question. 16. I have reviewed a copy of the Commerce Department’s notes of my March 23, 2018 conversation, marked as page 1276 in the Administrative Record (“AR 1276”) for this case.

78228 • ACLU • APPENDIX part:66 NP 4:51 3/24/19 374533

17. AR 1276 states that “Ms. Pierce stated that including a question on citizenship could make people less likely to respond, but that there is no data to predict how much lower the response rate might be.” I do not recall making this statement as worded here. Any statement like this would have been in the context of stressing the importance of conducting specific tests for the purpose of predicting the response rates. Adding a citizenship question to the Decennial Census introduces risk specifically because the impacts have not been tested. 18. AR 1276 states that I “noted that in the only specific situation she was aware of that sensitive questions were tested on a short questionnaire, there was no impact on response rates.” I did not state that “in the only specific situation that I was aware of that sensitive questions were tested on a short questionnaire, that there was no impact on response rates.” However, I did discuss Nielsen’s use of certain ACS questions and how Nielsen has tested those questions specifically to understand any impact to response. I did not provide any written reports with testing results nor did I provide Nielsen data in an attempt to estimate the impact of adding a citizenship question to the Decennial Census. Executed on October 25, 2018. /s/ Christine Pierce

78228 • ACLU • APPENDIX part:66 NP 4:51 3/24/19 375534

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

Civil Action No. 1:18-cv-05025-JMF ______

NEW YORK IMMIGRATION COALITION, CASA DE MARYLAND, AMERICAN-ARAB ANTI-DISCRIMINATION COMMITTEE, ADC RESEARCH INSTITUTE, and MAKE THE ROAD NEW YORK, Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE; and WILBUR L. ROSS, JR., in his official capacity as Secretary of Commerce, and

BUREAU OF THE CENSUS, an agency within the United States Department of Commerce; and RON S. JARMIN, in his capacity as performing the non- exclusive functions and duties of the Director of the U.S. Census Bureau, Defendants.

______

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 376535

DECLARATION OF ELIZABETH PLUM

Elizabeth Plum, pursuant to the provisions of 28 U.S.C. § 1746, declares under penalty of perjury as follows: 1. I am the Vice-President of Policy for Plaintiff the New York Immigration Coalition (“NYIC”). In that capacity, I am responsible in part for NYIC’s education and outreach efforts around the 2020 Census. I am also one of the NYIC executives responsible for the organization’s budgeting, fundraising, and policy priorities. I have been on staff at NYIC for over five years. 2. NYIC is an umbrella policy and advocacy organization for nearly 200 groups in New York State, representing the collective interests of New York’s diverse immigrant communities and organizations. NYIC is headquartered at 131 West 33rd St, New York, NY 10001. 3. NYIC’s mission is to unite immigrants, members, and allies so that all New Yorkers can thrive. NYIC envisions a New York State that is stronger because all people are welcome, treated fairly, and given the chance to pursue their dreams. NYIC pursues solutions to advance the interests of New York’s diverse immigrant communities and advocates for laws, policies, and programs that lead to justice and opportunity for all immigrant groups. It seeks to build the power of immigrants and the organizations that serve them to ensure their sustainability, improve people’s lives, and strengthen New York State.

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 377536

4. NYIC’s nearly 200 members are dues-paying nonprofit organizations that are committed to advancing work on immigrant justice, empowerment, and integration. NYIC’s member organizations—located throughout New York State and beyond—all share NYIC’s mission to serve and empower immigrant communities. NYIC’s members include grassroots community groups; social services providers; large-scale labor and academic institutions; and organizations working in economic, social, and racial justice. Representatives of NYIC’s member organizations serve on the NYIC Board of Directors. 5. Many of NYIC’s member organizations receive funding from a variety of local, state, and federal government sources to carry out social service, health, and education programs. Many of the programs from which NYIC member organizations receive money allocate funding based on population and demographic data generated by the Decennial Census. Among other Census- guided programs, NYIC member organizations receive funding through the Medical Assistance Program, also known as Medicaid; the Special Supplemental Nutrition Program for Women, Infants, and Children; the State Children’s Health Insurance Program; programs authorized under the Workforce Innovation and Opportunity Act; English Language Acquisition Grants; the Corporation for Community & National Service , which operates the AmeriCorps program; and formula grants authorized by the Violence Against Women Act.

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 378537

6. For example, several NYIC members, including Chhaya Community Development Corporation (“Chhaya’’), receive funding through the Community Development Block Grant program. 7. For another example, NYIC members Family Health Centers at NYU Langone, Little Sisters of the Assumption Family Health Services (“LSA”), Korean Community Services of Metropolitan New York, and Planned Parenthood of New York City receive funding through Medicaid to provide community health services. 8. For a third example, NYIC member Chinese- American Planning Council (“CPC”) receives funding through the Workforce Innovation and Opportunity Act to provide education, training, and other services to aid job seekers in securing employment. 9. NYIC also receives the value of $799,109 in Census-guided funding from the Corporation for National & Community Service for 21 positions filled by AmeriCorps VISTA members and approximately $20,000 for travel and administrative costs related to those positions. Through this program, NYIC has placed 17 AmeriCorps VISTA members into positions at member organizations, while 4 AmeriCorps VISTA members work directly with NYIC. 10. As an organization, NYIC has an ongoing commitment to promoting engagement in the Decennial Census among individuals served by its member organizations—in large part because of how critical the population count determined by

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 379538

the Census is for our members’ level of governmental funding and their level of political representation at the federal, state, and local levels. For example, NYIC partnered with the New York Community Media Alliance to launch an outreach campaign to boost immigrant participation in the 2010 Census. As part of that effort, NYIC coordinated public service announcements in 24 languages that appeared in 69 newspapers. NYIC also held press briefings with elected officials. These efforts helped to increase New York City’ s mail-in 2010 Census participation rate by approximately 3%. 11. For the 2020 Census, NYIC has already begun its outreach efforts. Since the beginning of 2018, it has helped form “New York Counts 2020,” a growing, non-partisan coalition of more than 50 diverse organizational stakeholders across New York to advocate for a fair and complete enumeration. This broad-based coalition, which was formally launched in March 2018, is composed of racial, ethnic, immigrant, religious, health, education, labor, housing, social services, and business groups working in partnership with state and local government officials. 12. NYIC is investing resources to solidify the work and reach of New York Counts 2020 through robust advocacy, outreach, and mass educational forums. It has already begun disseminating online petitions, petitioning Community Boards to pass resolutions for a fair and accurate count, and co- convened an all-day statewide conference, “Making New York Count in 2020.” NYIC will continue coordinating the working committees of

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 380539

New York Counts 2020, including by: coordinating “train the trainer” sessions throughout the state to equip leaders with tools to educate their communities on the importance of the Census; devising effective messaging to convince hard-to- reach communities to participate; empowering coalition members to assist their communities in completing the Census online; and advocating to ensure that there are no unnecessary barriers impeding marginalized communities from being counted while also ensuring their privacy is protected. 13. In its already extensive 2020 Census outreach, NYIC has faced, and wi11 continue to face, a more difficult Census-response environment. NYIC was already facing more significant challenges in its Census outreach even before the decision to institute the citizenship question because New York immigrant communities are more skeptical and fearful of interacting with the government due to the Trump Administration’s persistently hostile and discriminatory actions and attitudes toward immigrants of color. Now, New York immigrant communities’ heightened fear of interacting with government workers has increased even further due to the decision to add the citizenship question. This fear extends not only to undocumented immigrants or non-citizens with legal status, but also to family and household members of noncitizens who will be concerned that participating might endanger their loved ones. 14. The decision to add a citizenship question to the 2020 Decennial Census has required NYIC to

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 381540

make substantial and additional investments to achieve Census participation rates comparable to what we what would have achieved absent this decision. Over the next three years, NYIC is planning to spend approximately $1 million on community education and outreach efforts to work towards a complete and accurate count within the communities that NYIC and its member organizations serve. To the best of my knowledge, the $1 million that NYIC anticipates spending on Census-related work represents an increase of approximately 60% over what the organization would have spent in the absence of a citizenship question. 15. This additional spending is partially a response to reports NYIC has received from its member organizations that some of their members have expressed an unwillingness to participate in the Census as a result of the citizenship question. NYIC has heard such reports from many of its member organizations, including CPC, Arab- American Association of New York (“AAANY”), Masa, Chinese Progressive Association, MinKwon Center for Community Action, and Chhaya. 16. For example, in CPC’s Childhood Development Programs, parents who are immigrants of color have been asking CPC staff members whether or not it is “the law” to fill out the Census form and if or how they would be penalized if they chose not to respond. They also ask what will happen if they don’t answer a question during an in-person visit. Some have expressed concern that they may receive a visit from Immigration and Customs Enforcement if

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 382541

they indicate that there are noncitizens in their household. They have expressed reluctance to participate in the Census at all. In CPC’s Senior Centers, some seniors who are immigrants of color have been asking whether an incomplete mail-in form would be enough to trigger an in-person visit from ICE or another government official. In CPC’s Brooklyn Community Center, the senior services program has received an unusually high number of inquiries from community members on how to apply for citizenship. Some seniors and some younger generation Asian-Americans have also expressed concern because they are aware that Census data was used to identify Japanese- Americans and place them in internment camps during World War II. 17. For another example, AAANY has reported that many members of the Arab and Arab- American community they serve in and around the Bay Ridge section of Brooklyn have been reaching out to ask if they can refuse to fill out the Census because those community members are worried about the impact it will have on them or members of their households who have uncertain legal status. AAANY’s Adult Education Program Manager Somia Elrowmeim shared the fearful sentiments expressed by her students, many of whom said that they were apprehensive about having their names disclosed because of their legal status. AAANY’s adult education program consists of 110 women who are recent immigrants, primarily from Yemen and Egypt and between the ages of 18-68. According to Ms. Elrowmeim:

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 383542

“A lot of the women who I work with do not feel comfortable if someone asks them if they are a citizen. They feel like they will be targeted and treated like second- class citizens. Usually, if they see a question like that on any application they usually do not fill it out. Especially under this administration, the Arab community does not feel comfortable to say if they are a citizen or even a green card holder. Many struggle and endure many difficulties to become citizens and at the end they are still being asked where they are from. Our students and women have expressed how they feel like something is wrong and that they are being targeted with these questions. My members do not want their name out there, they are not comfortable because they are not citizens yet and do not feel like they have enough protections.” 18. For a third example, Chhaya has reported to NYIC that they have heard from members of the South Asian community they serve that the citizenship question is very concerning to them and they do not trust the federal government with such information. Those community members have expressed hesitation about participating in the Census specifically because of the citizenship question, including one community member who said at Chhaya’s monthly community meeting in May 2018 that if a citizenship question is added to the Census she will not fill out the Census, and will advise her family to do the same. She

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 384543

expressed that she felt scared that the information would be used to target members of her community for adverse immigration consequences. Chhaya has reported that they anticipate lower than average Census participation among their community members in 2020 if the citizenship question is included. Chhaya has reported that the fear generated by the citizenship question will require Chhaya to conduct more extensive education and outreach to the South Asian community regarding the 2020 Census than Chhaya would have conducted otherwise. 19. For a fourth example, Desis Rising Up and Moving, a member of the New Yark Immigration Coalition, reported that a man who emigrated to the United States from Guyana in the 1970s said that even as a citizen, he would not complete the Census if it contains a citizenship question because many people in his family and community are not citizens. This individual believed that this information could be used to target his family members who are non-citizens for deportation based on his understanding that Census data was used to target Japanese-American citizens for internment camps during the second World War. 30. For a fifth example, Little Sisters of the Assumption Family Health Service, a community- based organization and health services provider in East Harlem and a member of the New York Immigration Coalition, reported to NYIC a conversation with a community member about the Census and the fear about what a citizenship question could mean for his family and the

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 385544

community they serve. The community member is a 30-year-old male resident of East Harlem and is originally from Mexico. He is a DACA recipient and lives in a household with persons of mixed immigration status. He reported that he was afraid to answer a citizenship question on the Census because he is concerned that the Trump Administration may use that data to hurt his family and community. 21. NYIC has also received these reports directly from immigrants, including one from an immigrant from India who told an NYIC staff member that he would not fully participate in the Census because he is still trying to build a life within his community in New York and, that after everything he has had to sacrifice to live in this city and country, he cannot afford to put himself in danger of deportation. 22. One of the central concerns of NYIC and its members organizations with respect to the citizenship question is the loss of privacy that members of the immigrant communities we serve would suffer by the publication of citizenship data on the neighborhood or even city block level. With citizenship data at that level of granularity available, especially when combined with similarly granular data on race and Hispanic origin, it would be possible to determine where there are a high level of non-citizens of color, and for law enforcement agencies, particularly immigration authorities, to engage in the profiling of immigrant communities of color. 23. Because of the heightened fear and suspicion created by the citizenship question, NYIC and its

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 386545

member organizations will be forced to expend more resources on their outreach efforts to try to reduce the effect of this question on the response rate in the immigrant communities they serve. Due to this strain on resources, NYIC is already fundraising to try to support its 2020 Census work. NYIC will need to apply for additional grants to sustain the increased need for 2020 Census outreach, further diverting its resources that would otherwise be spent on trying to obtain grants for other areas. Further, NYIC has already, and will continue to, divert resources from its other organizational priorities, including its work on health care and language access issues. For example, NYIC was undertaking a study and publication on adult English literacy and workforce development, which examines the critical role of English language acquisition in integrating immigrants into the workforce and preparing them to earn higher wages; however, that project has been postponed indefinitely because of the resources required to perform additional Census outreach and education work. Additionally, based on information reported by the Communications Committee of New York Counts 2020, NYIC and some of its member organizations will have to divert resources that would have been spent on education and outreach efforts to increase Census response rates among immigrant communities of color towards addressing the heightened fear generated by the citizenship question.

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 387546

I, Elizabeth Plum, declare under penalty of perjury that the foregoing is true and correct to the best of my knowledge and recollections. Dated: New York, New York October 26, 2018 /s/ Elizabeth Plum

78228 • ACLU • APPENDIX part: 67 NP 4:25 3/24/19 388547

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______

Civil Action No. 1:18-cv-2921 (JMF) ______

STATE OF NEW YORK, et al., Plaintiffs, —v.—

UNITED STATES DEPARTMENT OF COMMERCE., et al., Defendants.

______

AFFIDAVIT OF JACQUELINE TIEMA-MASSIE

Pursuant to 28 U.S.C. § 1746(2), I, Jacqueline Tiema-Massie, hereby declare as follows: 1. My name is Jacqueline Tiema-Massie. I am over the age of eighteen and have personal knowledge of all the facts stated herein. I make this Declaration in connection with State of New York, et al., v. United States Department of Commerce, et al.

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 389548

2. I am the Director of Planning, Research and Development as well as the Director of Grants Management for the Chicago Department of Family and Support Services (“DFSS”). I have served in this position since December 2013. I direct the preparation of complex federal, state, local, and private human service grants worth over $315 million annually, which represents more than 80% of DFSS’s budget. Additionally, I supervise a team of grant project managers and planners that are responsible for the development and submission of the Department’s 30 reoccur- ring grants; for identifying new funding sources to support DFSS programs and new initiatives; preparing key reports, plans, and assessments associated with DFSS grants; and for the project management of the Community Services Block Grant (“CSBG”) program ($11.1 million). 3. Prior to DFSS, I served as the Director of Planning, Research and Development for the Chicago Department of Public Health, Bureau of Public Health Preparedness and Emergency Response, where I was responsible for the development and submission of the Public Health Emergency Preparedness Grant ($10 million) and the Hospital Preparedness Program Grant ($3 million) programs. I was also responsible for emergency preparedness planning, performance management, and planning activities during an emergency preparedness response. 4. In regards to the City of Chicago budget, I am involved with the Department’s preparation and submission of the annual appropriation ordinance, specifically the 925 – Grant Funds

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 390549

table, which includes all anticipated DFSS grants for the upcoming fiscal year, including carryover funds. For 2018, the total DFSS 2018 anticipated funds ($309,419,194) and carryover funds ($6,862,000) totaled $316,281,194. My team is responsible for securing these funds to support both the Department’s direct services, performed through the City’s Community Service Centers and Senior Regional and Satellite Centers, and contracted delegate agency services that support citywide human service delivery across multiple DFSS program models and service areas, including childcare, youth development, workforce, homeless, human, domestic violence, and senior services. 5. DFSS receives multiple federal and federal- state pass-through grants to support the City of Chicago’s human services delivery system. The following formula-driven grants are determined, in whole or at least in part, by population information collected by the Census: Federal grants – Community Development Block Grant; Emergency Solutions Grant; Federal•state pass• through grants – Area Plan on Aging Grants (federal and state); and the Community Services Block Grant. 6. For the programs identified above, information gathered during the decennial census is used by federal funding agencies to determine funding levels to states and local jurisdictions. These federally originated grants support communities based on census derived data that enumerates the total population counts and data about poverty, income, sex, and race/ethnicity.

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 391550

7. Approximate grant amounts are as follows: Community Development Block Grant (“CDBG”) – $25,000,000 (this is DFSS’s portion; the total for the City of Chicago is $72,100,000); Emergency Solutions Grant (“ESG”) – $6,500,000; Area Plan on Aging Grant (Federal) – $13,000,000; Area Plan on Aging Grant (State) – $8,000,000; and the Community Services Block Grant (“CSBG”) – $11,050,000. CDBG, ESG, and CSBG provide support for all age groups, and the Area Plan on Aging (federal and state) funding supports individuals aged 60 and over. 8. The CDBG program works to address a wide range of community needs. The City of Chicago receives annual allocations from CDBG on a formula basis, and this formula takes into account the current population of the City of Chicago as well as the number of Chicagoans in poverty. In 2017, these programs provided services for over 696,000 families citywide. 9. DFSS CDBG funds are used to operate homeless shelters and supportive service programs for people and families who are experiencing homelessness or are at imminent risk of homelessness so that they attain or maintain safe and secure housing; to provide emergency food supplies to at-risk populations; to conduct Intensive Case Advocacy for seniors which includes in-home assessments, case advocacy and support, and ongoing monitoring to vulnerable and socially isolated seniors and to provide home-delivered meals to seniors; to provide job readiness services, career counseling, vocational skills training, job placement assistance, and other workforce services;

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 392551

and to provide assistance and advocacy to those who have been victims of domestic violence, including counseling, case management, legal services, and supervised visitation and safe exchange. 10. The ESG program provides services to people and families experience homelessness. In 2017, ESG funding provided services for over 17,800 families throughout the City of Chicago. Locally, ESG funds are used to engage people and families experiencing homelessness; improve the number and quality of emergency shelters; support shelter operations; provide essential services to shelter residents; rapidly re-house people and families experiencing homelessness; and prevent families and individuals from becoming homeless. 11. The CSBG provides funds to alleviate the causes and conditions of poverty within communities. Locally, the Chicago Department of Family Support Services (DFSS) leads the charge to reduce poverty, in part, by providing CSBG services and supports to Chicago’s most vulnerable citizens. This is accomplished by delivering social services to residents through partnerships with a network of contracted/community-based (delegate) agencies and directly through DFSS. Through the CSBG contracted/community-based delegate agency network, DFSS provides the following services: Homeless Services, which provide support services for homeless persons, homeless street outreach, assessment, and counseling; and Workforce Services, which provide economic opportunities for low- income persons who are unemployed, job readiness services, career counseling, vocational skills

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 393552

training, job placement assistance, re-entry assistance, and other workforce services. Through directly administered programs DFSS provides the following services: DFSS Community Service Center (CSC) services, which provide client intervention and stabilization services, case work, case management, information and referral, emergency food referrals, transportation assistance, financial assistance program referrals and multiple co-located services (including veteran services, public benefits assessments, educational workshops, etc.); and The Scholarship Program, which provides post-secondary educational scholarships to CSBG eligible clients in an effort to achieve economic security and stability. In addition, the DFSS Community Service Centers provide emergency response to victims of disasters, such as people experiencing weather related problems. In instances of extreme (heat/cold) weather conditions, the facilities also serve as Warming and Cooling Centers to those in need. When the State of Illinois receives CSBG funds from the federal government, the state distributes 90% of these funds to the Community Action Agency (“CAA”) Network, which includes local governments such as the City of Chicago. The distribution of these funds to the Illinois CAA Network is based solely upon a formula utilizing a proportion of the state’s poverty rate, which is derived from the Census. In 2017, the DFSS CSBG programs, through DFSS direct and delegate agencies, provided services to over 26,600 families throughout Chicago.

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 394553

12. The Area Plan on Aging grants (federal and state) provide funding to maximize the quality of life of people aged 60 years and over and support a wide range of senior services including multi- purpose senior centers, nutritional aid (home delivered meals and congregate dining), and caregiver and ombudsman services. The City of Chicago has been designated as an Area Agency on Aging by the State of Illinois Department on Aging (“IDOA”) and is responsible for coordinating these community-based services to help those age 60 years and older stay as independent as possible in their homes and communities and avoid hospitalization or nursing home care. 13. The Area Agencies on Aging, including the City of Chicago, receive funding from IDOA based on a formula which takes into consideration the number of older citizens and elderly minorities in the area, as well as the number of seniors living in poverty, in rural areas, and alone. In order for a particular factor to be included in the funding formula, it must, among other things, be based on data which is derivable from the Census. Specifically, the funding formula factors are weighted as follows: the number of the state’s population age 60 years and older comprises 41% of the total; the number of the state’s population age 60 years and older, at or below the poverty threshold constitutes 25% of the total; the number of the state’s population age 60 years and older residing in rural areas comprises 9% of the total; and the final 25% is based on the state’s elderly population evidencing indicators of greatest social need. This final 25% is amalgamated as follows:

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 395554

those who are at least 60 years of age and a member of a minority group make up 10%; those at least 60 years of age and who live alone constitute 7.5%; and the number of the state’s population age 75 years and over comprise 7.5%. In 2017, the DFSS Area Plan on Aging grants served over 230,000 seniors citywide. 14. An accurate count of the populations served through the aforementioned programs is extremely important to serving the basic human needs of these populations, including housing, food, education, employment and income, and other needs. Many of these populations are extremely vulnerable and are from some of Chicago’s poorest neighborhoods. 15. Reducing funding to any of these programs puts these populations at greater risk to a wide range of vulnerabilities such as increased poverty, homelessness, substandard housing, hunger/food insecurity, unemployment, underemployment, lack of job skills training, social isolation, lack of access to higher education, exposure to intimate partner violence, and exposure to extreme weather. Reduced funding will also negatively impact DFSS’s contracted/community-based delegate agencies and their staff that provide these services and rely on this funding to successfully operate and manage their community programs. Based on similar analyses I have conducted in the past to address state budget concerns, these outcomes are very likely to occur as a result of reduced funding. 16. DFSS receives multiple federal, federal-stat e pass-through, state, and local funding to support its human services delivery in the City of Chicago.

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 396555

Some of DFSS’s largest federal and federal-state pass-through grants, including CDBG, ESG, CSBG, and Area Plan on Aging funding would significantly be impacted as a result of a reduction in funding due to a census miscount. 17. In addition to the service population being affected, DFSS administrative and program staff associated with this funding would be impacted. Delegate agencies which heavily rely on government support to operate programs, including their staff and operations, and the communities in which these organizations are located would all be negatively impacted. 18. Additionally, DFSS’s direct services would be impacted. For example, a reduction in funding to the CSBG would significantly impact low-income families. CSBG supports, among many other things, the operation of six DFSS Community Service Centers to assist individuals and families in need to access a wide range of resources, from shelter, food, and clothing to domestic violence assistance, drug rehabilitation, job training, and prison re-entry services. The Area Plan on Aging grants support the operations of six DFSS Regional and 15 Satellite Senior Centers. DFSS Senior Centers offer a variety of social, educational, cultural, and recreational activities for seniors and their informal caregivers. Many seniors utilize center resources for life enrichment activities, computer learning classes, internet access, fitness programming, congregate dining, caregiver services, and much more. Any reduction in funding could negatively impact these DFSS

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 397556

Centers that are integral to many Chicago neighborhoods, persons and families. 19. Chicago is a very diverse and multi-ethnic city, with 560,000 foreign-born residents from over 140 countries and more than 100 languages spoken. A strong and accurate census count is extremely important to ensuring that the populations served by federal and federal-state pass- through funding that the City receives will continue to receive support for the vital services that meet their basic human needs, such as food, shelter, quality child care, employment and training, safety planning, and other services as described above. A Census miscount would directly reduce and negatively impact the grant funding that the City receives. 20. The formula-driven grants that DFSS receives are vital and instrumental in providing direct and contracted services to hundreds of community organizations that provide a wide range of services to all Chicago residents, especially the most vulnerable, including immigrants. With such a substantial amount of federal dollars determined by Census data, multiple levels of the local safety net would be impacted. This could extend to other areas such as access to critical medical care, cash and nutrition assistance programs provided by state agencies through federal formula funding, as well as local housing assistance programs. This could further intensify Chicago residents’ need for local services in response to crises, while a reduction in funding would directly impact DFSS and its contracted partners’ ability to provide assistance to all Chicago residents. Any reduction

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19 398557

in funding would be devastating to the local human services infrastructure – to clients, nonprofit and community-based organizations, and most importantly to the at-risk populations that are served. I declare under penalty of perjury that, to the best of my knowledge, the foregoing is true and correct. Executed on this 23 day of October, 2018 /s/ Jacqueline Tiema-Massie Director of Planning, Research and Development Director of Grants Management Chicago Department of Family & Support Services

78228 • ACLU • APPENDIX part:68 NP 4:00 3/24/19